1. technology and computing
2. consumer electronics
3. telephones
4. mobile phones
5. smart phones

Mälardalen
Mälardalen University
University Press
Press Dissertations
Dissertations
No.
175
No. 175
PV WATER PUMPING SYSTEMS FOR
AGRICULTURAL APPLICATIONS
Pietro Elia Campana
2015
School
School of
of Business,
Business, Society
Society and
and Engineering
Engineering
Copyright © Pietro Elia Campana, 2015
ISBN 978-91-7485-189-2
ISSN 1651-4238
Printed by Arkitektkopia, Västerås, Sweden
Mälardalen University Press Dissertations
No. 175
Mälardalen University Press Dissertations
No. 175
PV WATER PUMPING SYSTEMS FOR AGRICULTURAL APPLICATIONS
Pietro Elia Campana
PV WATER PUMPING SYSTEMS FOR AGRICULTURAL APPLICATIONS
Pietro Elia Campana
Akademisk avhandling
som för avläggande av i energi- och miljöteknik vid Akademin för
ekonomi, samhälle och teknik kommer att offentligen försvaras fredagen
den 24 april 2015, 09.15 i Delta, Mälardalens högskola, Västerås.
Akademisk avhandling
Fakultetsopponent: Lawrence Kazmerski, University of Colorado Boulder
som för avläggande av i energi- och miljöteknik vid Akademin för
ekonomi, samhälle och teknik kommer att offentligen försvaras fredagen
den 24 april 2015, 09.15 i Delta, Mälardalens högskola, Västerås.
Fakultetsopponent: Lawrence Kazmerski, University of Colorado Boulder
Akademin för ekonomi, samhälle och teknik
Akademin för ekonomi, samhälle och teknik
Abstract
Grassland and farmland degradation is considered as one of the worst environmental and economic
threats for China. The degradation process negatively affects food and water security, economy, society
and climate changes.
Photovoltaic water pumping (PVWP) technology for irrigation is an innovative and sustainable solution
to curb the grassland degradation. At the same time it can promote the conservation of farmland,
especially in remote areas of China. The combination of PVWP technology with water saving irrigation
techniques and sustainable management of the groundwater resources can lead to several benefits. These
include enhancing grassland productivity, halting wind and rainfall erosion, providing higher incomes
and better living conditions for farmers. This doctoral thesis aims to bridge the current knowledge gaps, optimize system implementation and
prevent system failures. This work represents thus a step forward to solve the current and future nexus
between energy, water and food security in China, using PVWP technology for irrigation.
Models for the dynamic simulations of PVWP systems, irrigation water requirements (IWR) and crop
response to water have been presented and integrated. Field measurements at a pilot PVWP system in
Inner Mongolia have been conducted to analyse the reliability of the models adopted. A revision of the
traditional design approaches and a new optimization procedure based on a genetic algorithm (GA)
have been proposed to guarantee the match between IWR and water supply, to minimize the system
failures and to maximize crop productivity and thus the PVWP system profitability and effectiveness.
Several economic analyses have been conducted to establish the most cost effective solution for
irrigation and to evaluate the project profitability. The possible benefits generated by the PVWP system
implementation have been highlighted, as well as the effects of the most sensitive parameters, such
as forage price and incentives. The results show that PVWP system represents the best technical and
economic solution to provide water for irrigation in the remote areas compared to other traditional
water pumping technologies. The environmental benefits have been also addressed, evaluating the
CO2 emissions saving achievable from the PVWP system operation. The assessment of the feasible and
optimal areas for implementing PVWP systems in China has been conducted using spatial analysis and
an optimization tool for the entire supply chain of forage production. The results show that the potentials
of PVWP systems in China are large. Nevertheless, the feasible and optimal locations are extremely
sensitive to several environmental and economic parameters such as forage IWR, groundwater depth,
and CO2 credits that need to be carefully taken into account in the planning process. Although this doctoral thesis has used China as case study, PVWP technology can be applied for
irrigation purposes all over the world both for off- and on-grid applications leading to several economic
and environmental benefits.
ISBN 978-91-7485-189-2
ISSN 1651-4238
M. Peters, Dayton Daily News
To my family
Acknowledgments
This doctoral thesis was conducted at the School of Business, Society and Engineering, Mälardalen University, Västerås, Sweden. I gratefully acknowledge the Swedish International Development Cooperation Agency (SIDA),
Swedish Agency for Economic and Regional Growth (Tillväxtverket), and
Future Energy Center at Mälardalen University for the financial support.
First and foremost, I am deeply grateful to my supervisor Prof. Jinyue Yan
for giving me the opportunity to work on this project, his continuous and invaluable guidance, support, motivation, and all the great experiences I had
during my doctoral studies. It has been an honour to be one of his Ph.D. students.
My deep and sincere gratitude goes to my co-supervisor Assoc. Prof. Hailong Li for his continuous suggestions, guidance and patience in reviewing
my papers. I can say that your patience is out of this world! Thanks also for
teaching me that I should never give up!
I am extremely grateful to my supervisors at the International Institute for
Applied Systems Analysis for their support and great experience: Dr. Florian
Kraxner, Dr. Ian McCallum, Prof. Junguo Liu, and Dr. Sylvain Leduc.
I would like to thank the reviewers of my doctoral thesis: Prof. Hong-xing
Yang, Dr. Konstantinos Kyprianidis, and Dr. Tao Ma for their time, interest,
helpful comments, and support. Special thanks go to Daniel Berg for helping
me with translation and some language editing, and Mikael Gustafsson for his
work with the thesis layout.
I would like to thank Prof. Björn Karlsson for reviewing my licentiate thesis and for having helped me to solve my solar problems during my doctoral
studies!
Many thanks go to Prof. Ruiqiang Zhang, Prof. Hong-xing Yang, Prof.
Jiahong Liu and Assoc. Prof. Gang Xiao for their invaluable hospitality, guidance, ideas, knowledge and support during my trips and studies in China.
I am also grateful for the support from Solibro AB and for the close collaboration with Teroc AB, especially for the hard work and enlightening discussions with Sven Ruin.
I owe my deepest gratitude to my former supervisor in Italy, Prof. Umberto
Desideri, who first suggested me to come here at Mälardalen University in
2008.
I want to thank my colleagues at Royal Institute of Technology, Alexander Olsson and Chi Zhang, and at Institute of Water Resources and Hy-
dropower Research, Jun Zhang. My deep and sincere gratitude goes also to all
my colleagues and friends at Mälardalen University for the pleasant moments
spent together! I want to give my special thanks to all the staff at the School
of Business, Society and Engineering, but in particular to Ann-Sofie Magnusson, Elizabeth Catellani, Ewa Falkenö, Pablo Camacho Sanhueza, Saywan
Jamal, and Yvonne Arlestrand-Lundgren.
I want to thank the good colleagues and friends that I have met at the International Institute for Applied Systems Analysis: Moonil, Piera and Xi. Jie,
thanks a lot for all your moral support!
My most warm and special thanks go to Lundh family for the nice moments
spent together and for being my family in Sweden. Thanks a lot Janne! Thanks
a lot also to all my friends in China, Italy and Sweden.
Finally, I feel very much indebted to my family, especially to my father and
my brother, and to someone who is watching over me from up there.
Summary
Grassland and farmland degradation is considered as one of the worst environmental and economic threats for China. The degradation process negatively affects food and water security, economy, society and climate changes.
Photovoltaic water pumping (PVWP) technology for irrigation is an innovative and sustainable solution to curb the grassland degradation. At the same
time it can promote the conservation of farmland, especially in remote areas
of China. The combination of PVWP technology with water saving irrigation
techniques and sustainable management of the groundwater resources shows
several benefits. These include enhancing grassland productivity, halting wind
and rainfall erosion, providing higher incomes and better living conditions for
farmers.
This doctoral thesis aims to bridge the current knowledge gaps, optimize
system implementation and prevent system failures. This work represents thus
a step forward to solve the current and future nexus between energy, water
and food security in China, using PVWP technology for irrigation.
Models for the dynamic simulations of PVWP systems, irrigation water requirement (IWR) and crop response to water have been presented and integrated. Field measurements at a pilot PVWP system in Inner Mongolia have
been conducted to analyse the reliability of the models adopted. A revision of
the traditional design approaches and a new optimization procedure based on
a genetic algorithm (GA) have been proposed to guarantee the matching between IWR and water supply, to minimize the system failures and to maximize
crop productivity and thus the PVWP system profitability and effectiveness.
Several economic analyses have been conducted to establish the most cost
effective solution for irrigation and to evaluate the project profitability. The
possible benefits generated by the PVWP system implementation have been
highlighted, as well as the effects of the most sensitive parameters, such as
forage price and incentives. The results show that PVWP system represents
the best technical and economic solution to provide water for irrigation in the
remote areas compared to other traditional water pumping technologies. The
environmental benefits have been also addressed, evaluating the CO2 emission
saving achievable from the PVWP system operation. The assessment of the
feasible and optimal areas for implementing PVWP systems in China has been
conducted using spatial analysis and an optimization tool for the entire supply
chain of forage production. The results show that the potentials of PVWP systems in China are large. Nevertheless, the feasible and optimal locations are
extremely sensitive to several environmental and economic parameters such
as forage IWR, groundwater depth, and CO2 credits that need to be carefully
taken into account in the planning process.
Although this doctoral thesis has used China as a case study, PVWP technology can be applied for irrigation purposes all over the world both for offand on-grid applications providing several economic and environmental benefits.
Sammanfattning
Nedbrytning av gräsområden och jordbruksområden anses vara ett av de
värsta miljömässiga och ekonomiska hoten för Kina. Nedbrytningsprocessen
medför negativ påverkan på mat- och vattensäkerhet, ekonomi, samhällsliv
och klimatförändringar.
Solcellsteknologi för pumpvatten (engelsk förkortning: PVWP) för konstbevattning är en innovativ och hållbar lösning för att hindra nedbrytningen av
gräsområden och samtidigt gynnar den bevarandet av jordbruksområden, särskilt inom avlägsna områden i Kina. PVWP-teknologi i kombination med vattenbesparande tekniker för konstbevattning och ett hållbart bruk av grundvattenresurserna kan medföra flera fördelar – såsom ökad gräsproduktion, en
minskning av vind- och nederbördserosionen och ökade inkomster och förbättrade levnadsvillkor för jordbrukarna.
Syftet med denna doktorsavhandling är att överbrygga rådande kunskapsluckor, optimera implementeringen av systemet och förhindra systemfel.
Detta arbete betecknar sålunda ett steg framåt i lösningen av det rådande och
framtida sambandet mellan energi-, vatten- och matsäkerhet i Kina, i nyttjandet av PVWP-teknologi.
Modeller för dynamiska simulationer av PVWP-system, behovet av vatten
för grödor (engelsk förkortning: IWR) och grödornas reaktion på vatten har
presenterats och integrerats. Fältexperiment vid ett PVWP-pilotsystem i Inre
Mongoliet i Kina har utförts för att analysera tillförlitligheten inom de valda
modellerna. En ny bearbetning av de traditionella tillvägagångssätten med avseende på utformning och en ny procedur för optimering, baserad på en genetisk algoritm (GA), har föreslagits i syfte att garantera anpassningen mellan
behoven av vatten till skördar och tillgången på vatten, att minimera systemfel
och för att maximera produktionen från skördarna och sålunda även maximera
PVWP-systemets lönsamhet och effektivitet.
Flera ekonomiska analyser har utförts för att fastställa de mest kostnadseffektiva lösningarna för konstbevattning och för att utvärdera projektets lönsamhet. De möjliga fördelarna som genereras genom implementeringen av
PVWP-systemet har belysts såväl som effekterna av de känsligaste parametrarna, såsom priserna på foder och subventioner. Resultaten visar att
PVWP-systemet utgör den bästa tekniska och ekonomiska lösningen för att
tillhandahålla vatten till konstbevattning i de avlägsna områdena, jämfört med
andra traditionella teknologier för vattenpumpning, såsom dieseldrivna vattenpumpar och vindkraftsdrivna vattenpumpar, på grund av ett höga och säkra
kapitalvärdet och korta perioder för att uppnå avkastning. De miljömässiga
fördelarna har också uppmärksammats genom utvärdering av i vilken grad
koldioxidutsläppen kan minska genom nyttjande av PVWP-system.
Bedömningen av de lämpliga och optimala områdena för implementering
av PVWP-system i Kina har utförts genom bruk av spatial statistisk analys
och ett optimeringsverktyg för hela distributionskedjan inom foderproduktionen. Resultaten visar att möjligheterna med PVWP-system i Kina är stora.
Likväl är de lämpliga och optimala lokaliseringarna extremt känsliga i relation
till flera miljömässiga och ekonomiska parametrar, såsom fodrets IWR,
grundvattendjup och utsläppsrätter för koldioxid. I planeringsprocessen behöver dessa parametrar noga tas med i beräkningen.
Fastän den aktuella doktorsavhandlingen har använt Kina som fallstudie
kan PVWP-teknologi tillämpas för konstbevattningssyften över hela världen,
både för applikationer som är fristående i relation till elnätet och för applikationer som är kopplade till elnätet, vilket leder till flera ekonomiska och miljömässiga fördelar.
List of papers
This doctoral thesis is based on the following papers, which are referred to in
the text by their Roman numerals:
I.
II.
III.
IV.
V.
P.E. Campana, H. Li, J. Yan, “Dynamic modelling of a PV pumping system with special consideration on water demand”, Applied Energy, vol.
112, p. 635–645, 2013.
P.E. Campana, A. Olsson, H. Li, J. Yan, “An economic analysis of photovoltaic water pumping irrigation systems”, International Journal of
Green Energy, 2015 (Article in press).
P.E. Campana, H. Li, J. Zhang, R. Zhang, J. Liu, J. Yan, “Economic optimization of photovoltaic water pumping systems for irrigation”, Energy
Conversion and Management, vol. 95, p. 32–41, 2015.
P.E. Campana, H. Li, J. Yan, “Techno-economic feasibility of the irrigation system for the grassland and farmland conservation in China: photovoltaic vs. wind power water pumping”, manuscript for journal consideration.
P.E. Campana, S. Leduc, M. Kim, J. Liu, F. Kraxner, I. McCallum, H. Li,
J. Yan, “Optimal grassland locations for sustainable photovoltaic water
pumping systems in China”, Proceedings of the 7th International Conference on Applied Energy (ICAE 2015), March 28–31 2015, Abu Dhabi,
United Arab Emirates.
Parts of this thesis were included in the Licentiate thesis “PV water pumping
systems for grassland and farmland conservation”. In particular, Paper I was
also included in the Licentiate thesis; Paper II was included in the Licentiate
thesis as a conference paper and subsequently improved and submitted to the
International Journal of Green Energy; first drafts of Paper III and Paper IV
were part of the appended papers in the Licentiate thesis. Paper III has been
subsequently improved and submitted to Energy Conversion and Management. Paper IV is currently a manuscript for journal consideration.
Licentiate thesis:
P.E. Campana, “PV water pumping systems for grassland and farmland conservation”, Mälardalen University Press, Licentiate thesis 172, ISBN 978-917485-127-4, 2013.
The author has also contributed to the following relevant publications which
do not constitute part of this doctoral thesis:
I.
II.
III.
IV.
V.
VI.
VII.
VIII.
IX.
A. Olsson, P.E. Campana, M. Lind, J. Yan, “PV water pumping for carbon
sequestration in dry land agriculture”, Energy Conversion and Management, DOI:10.1016/j.enconman.2014.12.056, 2015 (Article in press).
A. Olsson, P.E. Campana, M. Lind, J. Yan, “Potential for carbon sequestration and mitigation of climate change by irrigation of grasslands”, Applied Energy, vol. 136, p. 1145–1154, 2014.
J. Zhang, P.E. Campana, J. Liu, R. Zhang, J. Yan, “Model of evapotranspiration and groundwater level based on photovoltaic water pumping
system”, Applied Energy, vol. 136, p. 1132–1137, 2014.
J. Yan, Z. Gao, P.E. Campana, A. Olsson, J. Liu, G. Yu, C. Zhang, J.
Yang, H. Li, “Demonstration and scale-up of photovoltaic water pumping
for the conservation of grassland and farmland in China”, SIDA report,
2014.
U. Desideri, P.E. Campana, “Analysis and comparison between a concentrating solar and a photovoltaic power plant”, Applied Energy, vol. 113,
p. 422–433, 2014.
H. Li, J. Yan, P.E. Campana, “Feasibility of integrating solar energy into
a power plant with amine-based chemical absorption for CO2 capture”,
International Journal of Greenhouse Gas Control, vol. 9, p. 272–280,
2012.
P.E. Campana, A. Olsson, C. Zhang, S. Berretta, H. Li, J. Yan, “On-grid
photovoltaic water pumping systems for agricultural purposes: comparison of the potential benefits under three different incentive schemes”, Proceedings of the 13th World Renewable Energy Congress (WREC XIII),
August 3–8 2014, London, United Kingdom.
P.E. Campana, Y. Zhu, E. Brugiati, H. Li, J. Yan, “PV water pumping for
irrigation equipped with a novel control system for water savings”, Proceedings of the 6th International Conference on Applied Energy (ICAE
2014), May 30–June 2 2014, Taipei, Taiwan.
C.L. Azimoh, B. Karlsson, F. Wallin, P. Klintenberg, S.P. Chowdhury, S.
Chowdhury, P.E. Campana, “The energy loss in guiding against equipment theft in Thlatlaganya Village, South Africa”, Proceedings of the 5th
International Conference on Applied Energy (ICAE 2013), July 1–4 2013
Pretoria, South Africa.
Contents
1
1.1
1.2
1.3
1.4
1.5
INTRODUCTION ....................................................................................... 1
Background ........................................................................................... 1
Knowledge gaps and challenges ........................................................... 2
Scope and objectives ............................................................................ 3
Contribution to knowledge ................................................................... 4
Thesis structure ..................................................................................... 5
2
LITERATURE REVIEW .............................................................................. 6
2.1 Grassland and farmland desertification: causes and measures ............. 6
2.2 Standalone water pumping systems for irrigation ................................ 8
2.3 PV water pumping systems for grassland rehabilitation and farmland
conservation ........................................................................................ 10
2.4 Summary ............................................................................................. 11
3
METHODOLOGY AND DESCRIPTION OF MODELS ................................... 12
3.1 PVWP system design and simulation ................................................. 13
3.1.1 Design of PVWP and WPWP systems .......................................... 14
3.1.2 Simulation of PVWP systems ....................................................... 18
3.2 Economic assessment of PVWP systems ........................................... 22
3.3 Optimization of PVWP systems ......................................................... 24
3.4 Suitable and optimal locations for PVWP systems ............................ 26
4
RESULTS AND DISCUSSIONS .................................................................. 28
4.1 Field measurements ............................................................................ 28
4.2 Impacts of irrigation water requirement on system design ................. 31
4.3 Optimization of integrated PVWP systems ........................................ 35
4.4 Technical and economic feasibility of PVWP systems ...................... 39
4.4.1 Comparison of standalone water pumping technologies ............... 39
4.4.2 Economic profitability ................................................................... 44
4.5 Suitable and optimal locations for PVWP systems ............................ 48
4.5.1 PVWP system design maps ........................................................... 51
5
SUMMARY OF PAPERS ........................................................................... 53
5.1 Paper I ................................................................................................. 53
5.1.1 Division of the work ...................................................................... 53
5.1.2 Results and discussions ................................................................. 53
5.2 Paper II ............................................................................................... 54
5.2.1 Division of the work ...................................................................... 54
5.2.2 Results and discussions ................................................................. 54
5.3 Paper III .............................................................................................. 54
5.3.1 Division of the work ...................................................................... 54
5.3.2 Results and discussions ................................................................. 55
5.4 Paper IV .............................................................................................. 55
5.4.1 Division of the work ...................................................................... 55
5.4.2 Results and discussions ................................................................. 55
5.5 Paper V ............................................................................................... 56
5.5.1 Division of the work ...................................................................... 56
5.5.2 Results and discussions ................................................................. 56
6
CONCLUSIONS ....................................................................................... 57
7
LIMITATIONS AND FUTURE WORKS ....................................................... 58
REFERENCES ................................................................................................. 59
PUBLICATIONS .............................................................................................. 65
List of figures
Figure 1: Relationship between scope, research questions, objectives, and
appended papers. .......................................................................... 4
Figure 2: Difference between fenced and overgrazed grassland
productivity in Inner Mongolia, China. ........................................ 7
Figure 3: Methodological approach. .......................................................... 12
Figure 4: Models for the design and simulation of PVWP systems for
irrigation. .................................................................................... 14
Figure 5: Dynamic performances of the centrifugal pump and inverter. ... 20
Figure 6: Initial capital cost (ICC) of PVWP systems and PVWP
components as a function of the PVWP capacity. ...................... 25
Figure 7: Schematic diagram of the conducted field measurements.......... 29
Figure 8: Measured and modelled hourly water flow vs. power input. ..... 30
Figure 9: Measured and modelled hourly reference evapotranspiration. ... 30
Figure 10: Measured and modelled hourly groundwater level. ................... 31
Figure 11: Irrigation water requirement (IWR) and design ratio for PVWP
systems in Hails, Inner Mongolia, China. .................................. 32
Figure 12: Irrigation water requirement (IWR) for Alfalfa and pasture grass
in Gancha, Qinghai, China. ........................................................ 33
Figure 13: Current and future (2050 A2 IPCC scenario) trend of Alfalfa
irrigation water requirement (IWR) in Gancha, Qinghai,
China. ......................................................................................... 33
Figure 14: Pumped water and irrigation water requirement (IWR) for
different PVWP systems in Xining, Qinghai, China. ................. 34
Figure 15: Crop yield (Alfalfa) and pumped water volume as a function of
the PVWP system capacity. ........................................................ 35
Figure 16: Progress of the genetic algorithm (GA). ..................................... 36
Figure 17: Effect of tilt angle and azimuth angle on the annual revenues. .. 37
Figure 18: Well water level trend induced by the optimized system during
the irrigation season.................................................................... 38
Figure 19: Effect of forage price and PV module price on the optimal
PVWP systems size assuming no groundwater response
constraint. ................................................................................... 38
Figure 20: Initial capital cost (ICC) of the studied PVWP systems for
providing water to 1 ha Alfalfa in Xining, Qinghai, China. ....... 40
Figure 21: Variation of initial capital cost (ICC) and life cycle cost (LCC) of
PVWP and DWP systems between 1998 and 2013. ................... 41
Figure 22: Breakeven point analysis between PVWP and DWP systems. .. 41
Figure 23: Initial capital cost (ICC) of PVWP and WPWP systems as a
function of the solar irradiation and wind speed for the design
month, and specific costs of PV modules and wind turbine
(WT). .......................................................................................... 42
Figure 24: Mismatching between irrigation water requirement (IWR) and
pumped water from PVWP and WPWP systems on 10-days
period. ......................................................................................... 43
Figure 25: Daily dynamic simulations of the irrigation water requirement
(IWR) and pumped water by PVWP and WPWP systems in
May............................................................................................. 44
Figure 26: Net present value (NPV) and payback period (PBP) analysis. ... 45
Figure 27: Effect of the interest rate, and forage yield and price on the net
present value (NPV). .................................................................. 45
Figure 28: Payback period (PBP) analysis as a function of the forage price
and specific costs of PV modules. .............................................. 46
Figure 29: Effect of the interest rate, forage yield and price, PVWP systems
components, and additional revenues on the net present value
(NPV) (W stands for well construction, I for irrigation system, E
for surplus of electricity generation, R for renewable electricity
generation incentives, C for carbon credits). .............................. 47
Figure 30: Disagreement map of the suitable areas for PVWP systems
identified by the IIASA and ADB approach. ............................. 48
Figure 31: PVWP forage production as a function of the market forage price
for different studied scenarios. ................................................... 49
Figure 32: Selection of the optimal locations for forage market price 200
(left) and 400 (right) $/tonne DM. .............................................. 50
Figure 33: PVWP system design map for the irrigation of 1 ha Alfalfa in
Qinghai Province, China. ........................................................... 51
List of tables
Table 1:
Table 2:
Table 3:
Table 4:
Table 5:
Table 6:
Table 7:
AquaCrop parameters for Alfalfa yield simulation. ................... 22
Genetic algorithm (GA) parameters. .......................................... 24
Decision variables and corresponding design spaces. ................ 25
Spatial data used for the assessment of the technically suitable
grassland areas for PVWP systems for irrigation. ...................... 27
PVWP system components and characteristics. ......................... 28
Measurements carried out during the tests and the corresponding
instruments, resolutions, and accuracies..................................... 29
Existing and optimized PVWP systems characteristic
parameters. ................................................................................. 36
List of principal symbols and acronyms
AC
ADB
APV
c
CF
CWR
D
DC
DM
DWP
ea
es
Es
ETc
ETc,adj
ETo
Ew
f
f(v)
FAO
fm
fuelann
g
G
GA
Gb,t
Gd,h
Gd,t
Gg,h
Gg,t
Gr,t
Hirr
Ho
hp
Alternate current
Asian Development Bank
PV array area (m2)
Scale factor (m/s)
Cash flow ($)
Crop water requirement (mm or m3/ha per unit of time)
Hydraulic diffusivity (m2/hour)
Direct current
Dry matter
Diesel water pumping
Actual vapour pressure (kPa)
Saturation vapour pressure (kPa)
Monthly average daily solar irradiation (kWh/m2/day)
Evapotranspiration in cultural conditions (mm or m3/ha per unit of
time)
Actual evapotranspiration (mm or m3/ha per unit of time)
Reference evapotranspiration (mm or m3/ha per unit of time)
Specific monthly average daily energy yield of the wind turbine
(kWh/kWr/day)
System failure
Probability density function of the wind speed (%)
Food and Agriculture Organization
Matching factor
Annual fuel cost ($)
Acceleration due to gravity (9.8 m/s2)
Soil heat flux density (MJ/m2 per unit of time)
Genetic algorithm
Beam solar radiation on the tilted surface (kW/m2)
Diffuse horizontal solar radiation (kW/m2)
Diffuse solar radiation on the tilted surface (kW/m2)
Global horizontal solar radiation (kW/m2)
Global solar radiation on the tilted surface (kW/m2)
Reflected solar radiation on the tilted surface (kW/m2)
Required head to operate the irrigation system (m)
Operational hydraulic head (m)
Level of the pump measured from the static water level (m)
Hr
i
ICC
ICCann
IIASA
IPCC
ir
IRR
IWR
IWRHR
IWRj,m
IWRPA
IWRt,m
j
k
K0
Kc
Ks
KTH
Ky
LCC
LR
m
MPPT
N
n
NOCT
NPP
NPV
omrann
p
Pann
PBP
pc
Pe
pf
Pp,o
Pp,PVWP
PPV
Pr
Pr,WPWP
PRMSE
Reference hydraulic head (m)
Imaginary number
Initial capital cost ($)
Annualized initial capital cost ($)
International Institute for Applied Systems Analysis
Intergovernmental Panel on Climate Change
Interest rate (%)
Internal rate of return (%)
Irrigation water requirement (mm or m3/ha per unit of time)
Institute of Water Resources and Hydropower Research
Monthly average daily irrigation water requirement for the j-th
crop in the m-th month (m3/ha/day)
Institute of Water Resources for Pastoral Areas
Total monthly average daily irrigation water requirement in the mth month (m3/ha/day)
j-th irrigated crop
Weibull shape factor
Zero-order modified Bessel function
Cultural coefficient
Water stress coefficient
Kungliga Tekniska Högskolan (Royal Institute of Technology)
Yield response factor
Life cycle cost ($)
Leaching requirements (%)
m-th month
Maximum power point tracker
Project lifetime (year)
Year of the investment (year)
Nominal operating cell temperature (°C)
Net primary productivity
Net present value ($)
Annual operation, maintenance and replacement cost ($)
Velocity-power proportionality number
Annual profit ($)
Payback period (year)
Pumping cycle (hour)
Effective precipitation (mm or m3/ha per unit of time)
Specific forage price ($/tonne DM)
Operational pump power (kW)
Photovoltaic water pumping power peak (kWp)
PV array power output (kW)
Wind turbine rated power (kWr)
Wind power water pumping rated power (kWr)
Percentage root mean square error (%)
Pv
PV
PVWP
PW
Qo
Qr
r
Rann
Rn
S
s
sh
SH
smax
STC
T
t
Ta
Tcell
TDH
TSTC
v
v2
vi
Vmp
vo
Vp,d
vr
Vs,d
WPWP
WT
x
Ya
Ym
Power produced by the wind turbine at wind speed v (kW)
Photovoltaic
Photovoltaic water pumping
Present worth ($)
Operational water flow (m3/hour)
Reference water flow (m3/hour)
Distance from the pumping well (m)
Annual revenue ($)
Net radiation at the soil surface (MJ/m2 per unit of time)
Storativity
Drawdown (m)
Hourly drawdown (m)
Static head (m)
Maximum drawdown (m)
Standard test conditions
Aquifer transmissivity (m2/hour)
Time (hour)
Ambient temperature (°C)
Photovoltaic cell temperature (°C)
Total dynamic head (m)
Temperature at standard test conditions (25 °C)
Average wind speed (m/s)
Average wind speed at 2 m above the ground (m/s)
Cut-in wind speed (m/s)
Voltage at maximum power point (V)
Cut-out wind speed (m/s)
Daily pumped water volume (m3/day)
Rated wind speed (m/s)
Daily sustainable pumped water volume (m3/day)
Wind power water pumping
Wind turbine
Total number of irrigated crops
Actual yield (tonne DM/ha/year)
Maximum yield (tonne DM/ha/year)
Greek symbols
α
Solar altitude (˚)
αC
Temperature coefficient of the PV power (%/°C)
Tilt angle (°)
β
Psychrometric constant (kPa/°C)
γ
γs
Solar azimuth angle (°)
Surface azimuth angle (°)
γsu
Declination angle (°)
δ
Δ
Slope of saturation vapour pressure curve (kPa/°C)
ƞinv
ƞirr
ƞm,o
ƞm,r
ƞp
ƞPV
ƞPV,STC
θ
λ
μ
μVoc
ξ
ρ
ρg
φ
ωh
ωp
Efficiency of the inverter (%)
Efficiency of the irrigation system (%)
Efficiency of the motor at the operational working conditions (%)
Efficiency of the motor at the reference working conditions (%)
Efficiency of the pump (%)
Efficiency of the PV module (%)
Efficiency of the PV module at standard test conditions (%)
Angle of incidence (°)
Continuous head losses (m)
Temperature coefficient of the PV efficiency (1/°C)
Temperature coefficient of the open-circuit voltage (V/°C)
Concentrated head losses (m)
Water density (1000 kg/m3)
Ground reflectance (%)
Latitude (°)
Hour angle (°)
Pumping frequency (Hz)
1 Introduction
1.1 Background
Grassland covers an area of 4 million km2 in China, accounting for 41.7% of
the total national land area [1]. They play a key strategic role in the sustainable
development and food security of the country, sustaining more than 100 million heads of livestock [2]. In addition, grassland is essential for preserving
water resources, and influence the water cycle, especially with regard to the
regulation and formation of river runoff [3]. Furthermore, these ecosystems
represent an important carbon sink [4].
However, Chinese grassland is suffering from varying degrees of degradation, leading to desertification and reduced grassland yield and biodiversity
[5]. It has been estimated that grassland degradation in China affects the lives
of 400 million people and results in an economic loss of 8 billion US dollars
per year, while also posing a threat to the breeding of livestock [2]. Land degradation and desertification in China are not only restricted to grassland but
also affect farmland, causing crop yield reduction and aggravating the country’s food security challenges. Desertification in the farming-pastoral lands of
North China accounts for 60% of the country’s total desertified area [6].
The rehabilitation of degraded grassland and the preservation of farmland
relies, in most cases, on irrigation to achieve higher grass and crop yields [7].
This is a commonly used practice for improving the forage yield in the United
States [8], New Zealand [9], Australia [10], and North and East African countries [7].
However, grassland irrigation may bring with it pressures in terms of water
and energy security, especially because the food and energy demand of the
pastoral and agricultural sectors is also influenced by population growth and
climate change [11, 12]. The major technical obstacle to irrigation in remote
farming and pastoral areas is access to the electricity grid. In these areas, photovoltaic (PV) water pumping (PVWP) technologies, implemented with water
saving irrigation techniques and aimed at sustainable management of water
resources, represent a technically suitable and renewable solution to provide
electricity in off-grid areas for irrigation of grassland and farmland. The application of such systems in grassland produces several direct and indirect
benefits, improving grass coverage and consequently decreasing the driving
forces of its degradation (i.e., wind and rainfall erosion).
1
Improved grassland coverage can further positively affect the hydrologic
balance, such as by increasing the rate of rainfall that recharges groundwater
and by increasing carbon sequestration, thus contributing to climate change
mitigation [4]. Grassland and farmland irrigation can also increase forage and
crop yields, resulting in higher incomes and consequently better living conditions for farmers. PVWP systems can further contribute to the improvement
of living standards by supplying water for drinking purposes and electricity,
especially in rural and remote areas. The use of such systems, in combination
with water saving irrigation techniques (such as micro drip or sub surface irrigation), can lead to a substantial reduction in energy and water requirements
of the pastoral and agricultural sectors. PVWP is therefore a promising technology that can help to meet future energy and food demand in a sustainable
way.
1.2 Knowledge gaps and challenges
PVWP systems have been studied and installed for over 40 years in off-grid
applications [13], especially for drinking purposes. Nevertheless, the drastic
fall in prices of PV modules due to the rapid worldwide growth of the PV
market over past years has boosted research and development of these systems, encouraging greater system flexibility and larger and new applications
[14]. Most scientific research carried out in this field is focused on system
design, optimization of system components (such as power conditioning systems and performance improvement of the solar array), and technical and economic comparisons between PV and other standalone power sources.
However, there is a knowledge gap in the systematic optimization of energy
systems, when considering irrigation water requirement (IWR), water resource availability, and crop yield under different water supply conditions.
System failures and corresponding economic losses still occur due to the inadequacy of system integration with the environment. In most works done so
far, PVWP systems have been considered as independent electric devices,
without taking into consideration how the system is affected by the environment (e.g., IWR and water resources) and how the system affects the environment (e.g., pumped water and crop yield). Thus, the first two research questions are: how does the variation in IWR during the irrigation season affect PVWP systems design and effectiveness? (Q1); and, how can the
PVWP systems be optimized from an integrated system perspective?
(Q2).
There is also a knowledge gap in the systematic analysis of the technical,
environmental, and economic feasibility of PVWP systems for the specific
purpose of halting the degradation of grassland and for promoting sustainable
development of pastoral and farmland area. It is also essential to identify economic, technical, and environmental constraints that can negatively affect the
2
operation of PVWP systems. In addition, in most studies conducted so far, the
profitability of PVWP systems has been underestimated, even though this is
significant for investors. The third research question is: are PV water pumping systems technically, economically, and environmentally feasible for
grassland restoration and farmland conservation? (Q3).
The assessment of the most suitable and optimal areas for implementing
PVWP systems is another crucial aspect that needs to be investigated for wide
and sustainable applications of the technology. Previous studies have not
taken into account the availability of water resources in identifying technically
suitable grassland areas for implementing PVWP systems. Moreover, conducted optimization neglected all costs related to the forage supply chain and
to the co-benefits of implementing PVWP systems for grassland and farmland
conservation. The last research question addressed in this doctoral thesis is:
which are the most suitable and optimal areas for implementation of
PVWP systems in grassland areas for forage production and how can we
identify them? (Q4).
1.3 Scope and objectives
The overall scope of this thesis is to assess the technical, economic, and environmental suitability of PVWP technology for pastoral and agricultural purposes, specifically for restoration of degraded grassland and farmland conservation.
Corresponding to the above-listed research questions, specific objectives
include the following:




providing a better understanding of the matching between PVWP system water supply, IWR, water resources, and crop response to water
for the improvement of the system design and reliability (O1);
proposing a novel integrated optimization approach that takes into
consideration the availability and response of groundwater resources
to pumping, the effect of water supply on crop yield, the investment
cost of PVWP, and revenue related to crop yield (O2);
identifying the most effective and efficient solution in technical and
economic terms to provide water for irrigation in remote areas, analysing the negative environmental effects, and direct and indirect benefits (O3);
providing a reliable methodology to assess technically suitable and
optimal locations for implementing PVWP systems for irrigation
(O4).
3
A summary of thesis organization in terms of scope, research questions, and
objectives, and the corresponding relationships with appended papers is
shown in Figure 1.
China has been used as a case study for purposes of this work. Nevertheless, all methodologies applied and results achieved in this doctoral thesis
could potentially be replicated in other areas of the world where irrigation is
required for both pastoral and agricultural purposes.
Figure 1:
Relationship between scope, research questions, objectives,
and appended papers.
1.4 Contribution to knowledge
Corresponding to the appended papers, the main contributions of this doctoral
thesis consist of:

4
developing an integrated framework to design and optimize PVWP systems for irrigation. Special consideration is given to the matching between
IWR and PVWP system water supply, and to the effect of pumping on the
groundwater resources and crop yield response;

providing a detailed economic analysis of PVWP systems. A comprehensive analysis of the potential economic and environmental benefits of
PVWP technology is conducted to support its implementation;

developing a new approach to assess the suitable and optimal locations
for implementing PVWP systems in the Chinese grassland.
1.5 Thesis structure
This thesis consists of the following chapters:
Chapter 1 Introduction
To introduce the background information, knowledge gaps and
challenges, objectives, contributions and thesis outline.
Chapter 2 Literature review
To review the situation of grassland desertification and farmland
conservation in China and the work conducted regarding the development and application of PVWP to date.
Chapter 3 Methodology and description of models
To describe the methodology and models adopted for the dynamic
simulations, the optimization, and economic evaluation.
Chapter 4 Results and discussions
To present the main results achieved in the appended papers of this
thesis and highlight the main discussion points.
Chapter 5 Summary of papers
To summarize the results of the appended papers and highlight the
author’s and co-authors’ contributions.
Chapter 6 Conclusions
To highlight the main findings of this doctoral thesis.
Chapter 7 Limitations and future works
To define the limitations of this doctoral thesis and to introduce the
potential future research areas.
5
2 Literature review
This chapter provides a literature review of previous studies conducted on the
following topics: grassland and farmland desertification, standalone water
pumping systems, and PVWP systems for grassland and farmland irrigation
in China. Current research gaps in the field of PVWP systems for grassland
and farmland conservation are summarized at the end of the chapter.
2.1 Grassland and farmland desertification: causes and
measures
The term desertification refers to land degradation, and is considered to be the
reduction or loss of biodiversity and crop yield in arid, semiarid, and dry subhumid areas due to climatic and anthropogenic factors [15]. A significant proportion (27.3%) of the Chinese national land area is affected by desertification, with a yearly increase of 2460 km2 [2]. At the same time, population
growth, income growth, land use change, and nutrition changes are some of
the factors that are undermining the food security of China [11]. Grassland
and farmland degradation is, therefore, a high priority issue for the Chinese
political, economic, and environmental agenda and is drawing the interest of
the scientific community. Over the past 20 years, scientists have debated on
the main causes of desertification in China to determine the right measures for
halting it. There are two main schools of thought. The first attributes the main
cause to human activities (such as overgrazing, land use change, irrational irrigation, over cultivation, and pollutant discharges) [16, 17], whereas the second blames climate change (global warming, drought, water and wind erosion)
as a strong driving force of desertification processes [18]. Both sets of causes
produce a lack of vegetation coverage and biodiversity, deteriorating the already fragile environmental and ecological situation.
While preventing desertification includes integrated water and land management practices, complex engineering and technological innovation [19],
the main recognised method of halting this phenomenon (or land degradation,
more generally) is increasing or maintaining vegetation coverage [20]. The
most typical approaches for halting the progress of desertification rely on stopping erosion, such as by enriching soil with nutrients and through afforestation
and reforestation practices.
6
Figure 2:
Difference between fenced and overgrazed grassland
productivity in Inner Mongolia, China.
The main scope of these measures is to enhance vegetation coverage and to
create barriers to halt the progress of desertified areas [19]. Both afforestation
and reforestation projects have the drawback of high water requirements for
their establishment. Moreover, arboreal vegetation transpires large quantities
of water and its deep well-developed root system can draw water directly from
the ground, leading to a reduction in the water table. This explains why, in
some cases, afforestation and reforestation practices (and their related water
resources demand) are controversial [21].
Another typical solution to prevent desertification (in particular grassland
desertification caused by overgrazing) is to fence some areas and reduce the
number of livestock heads per unit of grassland; this leads to higher soil fertility, vegetation biomass, biodiversity, and water storage [22]. The effectiveness of livestock exclusion practices for restoring degraded grassland due to
overgrazing has been assessed, with the results showing that exclosures can
positively affect vegetation restoration, soil fertility improvement (especially
soil nitrogen and phosphorous), and soil erosion reduction. Figure 2 shows the
difference between fenced and overgrazed grassland productivity for a site in
Inner Mongolia, China. Nevertheless, although fenced areas have been recommended for grassland restoration from a technical viewpoint, this approach
can lead to several socioeconomic disadvantages. The most significant is that
of further reducing already low incomes of farmers, especially in remote farming-pastoral areas.
Recently, the application of PVWP systems for grassland irrigation in
China has attracted more and more attention from the scientific community as
7
an innovative solution for preventing desertification, while at the same time
supporting the sustainable development of farmland in China [23].
2.2 Standalone water pumping systems for irrigation
The main technological alternatives used for water pumping in off-grid areas
are PV, diesel, and wind-powered water pumping (PVWP, DWP, and WPWP)
systems. PVWP, DWP, and WPWP systems have been widely used to supply
water for drinking, livestock, and irrigation purposes in remote off-grid areas.
A PVWP system for irrigation typically consists of five main components: PV
array, power control unit, pumping system, storage unit, and irrigation system
[24]. Multiple configurations of system layout exist and several technical
component choices are available, depending on reliability, performance, and
economic aspects [25].
PV modules can be installed on a fixed array or on a sun tracking system.
Fixed PV arrays are cheap to install and require no maintenance, whereas the
sun tracking system is expensive and requires maintenance of moving components, but allows harvesting of 30–40% more solar irradiation than the fixed
system. The controller depends on the pump motor type, whether direct current (DC) or alternate current (AC). In the first case, the AC pump is connected
to the PV array through a DC/AC inverter that transforms the DC power produced by the PV modules to AC power used to drive the pump motor. If a DC
pump is chosen, then coupling is arranged with a DC/DC converter. Most of
the controllers available on the market are also equipped with a maximum
power point tracker (MPPT) that allows extraction of the maximum power
produced by the PV array. The controller can also be omitted in the so-called
directly coupled PVWP systems, typically adopted for small applications.
In most cases, water storage units are absent from PVWP irrigation systems
since IWR and pumped water are synchronized. Nevertheless, if there is a
need for interim storage of the water supply, a storage system can be included
in the PVWP configuration. In the case of the storage system being a battery
bank, a charge controller interfaces the PV module with the battery and the
power-conditioning unit. The simplest and most reliable PVWP system layout
for irrigation purposes involves a direct connection between the irrigation system and the pump through a filtering unit. Crop irrigation can be performed
using several methods such as, micro drip, sprinkler, and furrow irrigation.
Micro-irrigation has the highest level of efficiency (of up to 90%) compared
to sprinkler and furrow irrigation [26]. It is thus preferred as a water saving
technique, especially in areas where water resources are limited. It is clear that
its specific investment cost is higher than that of other irrigation techniques
and thus a benefit-cost ratio analysis should be carried out during the planning
process.
8
Several works have dealt with PVWP systems for irrigation, highlighting
how solar powered irrigation can be considered an attractive application of
renewable energy. Kelley et al. [27] analysed the technical and economic feasibility of PVWP systems for irrigation, concluding that there are no technological barriers for implementation of PVWP systems. The main disadvantage
of PVWP systems is still tied to the high price of PV modules. Cuadros et al.
[28] presented a procedure to design PVWP systems for drip irrigation of an
olive tree orchard in Spain, focusing on the assessment of IWR. Hamidat et
al. [29] developed a program to test the performance of PVWP systems for
irrigation in regions of the Sahara. The study concluded that PVWP systems
were suitable for crop irrigation in small-scale applications. Glasnovic and
Margeta [30] proposed a new optimization model of PVWP systems for irrigation. The objective function was to minimize PVWP system size, taking
into account constraints related to IWR and water availability.
There have been many studies comparing different water pumping systems
for irrigation, in particular PVWP versus DWP and WPWP systems. PVWP
systems present several advantages compared to traditional DWP systems
used for irrigation in rural areas. The operation of PVWP systems is independent from fossil fuels and so overcomes all related disadvantages: fuel availability, fuel price fluctuations, fuel and oil spills, exhaust gases, and greenhouse
gas emissions [31]. Other major advantages of PVWP systems are their high
reliability and flexibility, low maintenance and operation costs, and the absence of noise during their operation.
Although PVWP systems currently have a shorter life cycle cost (LCC)
compared to DWP systems, due to the high operation and maintenance cost of
diesel engines [32], DWP systems have a much lower initial capital cost
(ICC). For this reason, DWP systems have, for several years, been considered
to be the best techno-economic choice, especially when the decision making
process relies mostly on low ICC. Nevertheless, the decline in the price of PV
modules has altered this trend, boosting the PV market in general and making
PVWP systems more competitive.
Water pumping is also one of the typical applications of wind power.
WPWP systems can be coupled with the pump through mechanical or electrical transmission. As with the PVWP system, the wind turbine can either be
directly connected to the pumping system or can be connected through a
power-conditioning unit [33]. Especially for livestock watering and crop irrigation, PVWP systems offer better performance than WPWP systems in terms
of synchronizing IWR and water supply [13]. Bouzidi [34] compared PVWP
and WPWP systems from a technical and economic point of view for the purpose of providing drinking water in a specific location (Adrar in Algeria).
Dıaz-Mendez et al. [35] focused mainly on an economic comparison of PVWP
and WPWP systems for irrigation of greenhouses, analysing the impact of
several parameters on competitiveness, such as distance from the electric grid,
need for a water storage tank, and water elevation.
9
Recent research on PVWP systems has mainly focused on system modelling. There has been particular focus on modelling of (and improvements in)
electrical components, such as PV modules [36], inverters [37] and controllers
[38], and on modelling of PVWP system performance, such as water flow under various operating heads [39] and CO2 emissions mitigation potential [40].
2.3 PV water pumping systems for grassland
rehabilitation and farmland conservation
In 2011, the Asian Development Bank (ADB) started to investigate the implementation of PVWP systems for grassland protection, with installation of a 2
kWp pilot system in Qinghai province [23]. The project demonstrated the technical and economic feasibility of PVWP systems for preventing grassland degradation, farmland conservation, and poverty alleviation in rural areas. The
following aspects related to the implementation of PVWP systems in grassland were addressed: overexploitation of water resources, institutional and financial barriers, comparison with traditional off-grid water pumping systems,
and guidelines for identifying the most suitable locations. Suitability was evaluated through spatial analysis of land cover, terrain slope, rainfall, temperature, and sunshine hours. An attempt at system optimization was carried out,
taking into account the incremental benefit of irrigation and assessing the rate
of investment return in relation to precipitation. The authors concluded that
the highest economic benefits of PVWP system could be achieved in areas
characterized by 350–400 mm of precipitation.
Following the work of the ADB, efforts were extended to Inner Mongolia
and Xinjiang Province, with two further pilot PVWP systems. Research focused on the effect of irrigation and irrigation technologies on water resources,
integration of PVWP technology in pastoral and agricultural applications, and
environmental and economic benefits. The feasibility of grassland irrigation
in Inner Mongolia was studied by Xu et al. [41], who focused mainly on the
evaluation of IWR during different hydrologic years and on the effect on
groundwater resources. A positive balance between IWR and water resources
showed that water resources were not affected by pumping. Gao et al. [42]
analysed the effects of PVWP systems on groundwater table variations in
Qinghai and the economic profitability of the system, taking into consideration the sales of forage as revenue. Irrigated grassland showed an increase of
200% in productivity compared to non-irrigated areas. Moreover, when taking
into account overall investment costs and the local grass market price, the
PVWP system was shown to have a payback period of eight years, thus
demonstrating an excellent economic return. A study conducted by Gao et al.
[43] showed the effects of irrigation on grassland productivity and biodiversity in Xilamuren grassland, Inner Mongolia. Results indicated that, even
10
though irrigation can improve both grass production and biodiversity, the halting of irrigation then led to more and faster degradation in irrigated areas than
in areas that had not been irrigated at all. The need for further and systematic
monitoring research therefore became clear.
A novel business model, which can be applied to integrated PVWP systems
for grassland and farmland conservation, was proposed by Zhang and Yan
[44]; this included environmental co-benefits, agricultural products, and social
visualization of all benefits. Jun et al. [45] estimated the CO2 emissions of a
PVWP system from a life cycle perspective. The results showed that the technology has great potential for global warming mitigation, especially if used to
irrigate degraded grassland, increasing carbon sequestration.
2.4 Summary
Most studies conducted that focus on the use of PVWP systems for irrigation
lack systematic integration and optimization. Few studies consider the dynamic operation of PVWP systems. In particular, there are no previous detailed studies that consider the dynamic matching between water supply and
IWR, and related dynamic effects on groundwater levels and crop yields for
optimization purposes.
There is still a lack of general comparison between PVWP and WPWP systems, especially for irrigation. The application of PVWP technology for
preservation of grassland and conservation of farmland in China has been the
objective of a few studies, which are related to the present doctoral thesis.
However, previous studies have not addressed neither the manner in which
different forage crops affect the design of PVWP systems, nor the way in
which climate change can affect design. Moreover, not many studies have assessed the profitability of PVWP systems for grassland irrigation. Previous
studies considered only forage sales for a specific location with a specific forage yield as PVWP system revenue, thus failing to provide a comprehensive
analysis of profitability. There is thus the need to analyse PVWP system suitability, including all potential benefits. Previous assessments of suitable and
optimal locations for PVWP systems in China also disregarded spatial parameters, such as groundwater availability, groundwater depth, and potential forage yield; they also did not take into account costs and benefits across the
entire forage supply chain.
11
3 Methodology and description of models
The methodology adopted in this thesis is illustrated in Figure 3.
Figure 3:
Methodological approach.
PVWP system design is based on a thorough analysis of available solar irradiation, IWR, and the effect of climate change on IWR. This thesis also presents a general procedure to design both PVWP and WPWP systems that take
into account irrigation of multiple crops and future climatic scenarios. Several
commercial tools have been used to simulate the PVWP sytem, namely PVsyst
[46], Trnsys [47], and Matlab [48].
PVsyst is considered as one of the most comprehensive programs for designing and simulating PV systems. Unlike most other programs, it also includes the tools for PV water pumping applications. Trnsys is an extremely
flexible and intuitive graphically based software for dynamic simulations of
renewable energy systems. Matlab is a high-performance language for technical computing with several advantages compared to conventional computer
languages. Cropwat [49] and AquaCrop [50] are flexible and educative tools
used to calculate IWR and to simulate crop yield under different water supply
conditions. Cropwat and AquaCrop are freely distributed by the Food and Agriculture Organization (FAO) of the United Nations. Dynamic modelling of
12
the integrated system is used to validate the design process through the matching of IWR and PVWP system water supply. Simulations are used to evaluate
groundwater and crop yield response to pumping.
Field experiments were conducted at a pilot PVWP system installed in Inner Mongolia to gather experimental data for analysing the reliability of the
simulations results. Furthermore, results gathered during the field trip were
used for proposing a novel optimization procedure based on a genetic algorithm (GA) for PVWP irrigation systems. The optimization of the PVWP system was conducted using Solve XL, an add-in for Microsoft Excel that has the
capability to solve various optimization problems by implementing GA [51].
The user friendly interface and built-in help make the use of Solve XL extremely easy.
The economic aspects of the PVWP systems were also thoroughly investigated. ICC and LCC were used to compare different water pumping technologies, and net present value (NPV), internal rate of return (IRR), and payback
period (PBP) were used to evaluate the capacity to produce benefits. The environmental advantages of using PVWP systems were analysed in terms of
CO2 emission reduction. The assessment of technically suitable and optimal
areas for implementation of PVWP systems was conducted through spatial
analysis using ArcGIS [52], GAMS [53], and BeWhere (a techno-economic
model that determines optimal size and geographic distribution of bioenergy
production plants) [54]. ArcGIS is a geographical information system with the
advantage of several spatial analysis tools. BeWhere is a mixed linear integer
program written in GAMS, tailored for complex and large scale modelling
applications. Based on spatial analysis of groundwater depth, IWR, and solar
irradiation, PVWP system design maps were developed as a new rapid tool
for policy makers, consultants, and farmers willing to invest in PVWP technology for irrigation.
Alfalfa (Medicago sativa) was selected as a reference crop throughout the
thesis since it is one of the most widely used forage crops. China is a major
importer of Alfalfa, especially from the USA, Canada, and Mongolia, with
such imports being crucial in meeting the growing domestic demand for food
and milk [55].
3.1 PVWP system design and simulation
Figure 4 provides an overview of the models related to the functioning of an
integrated PVWP system. The integrated PVWP system model is decomposed
into five sub-models, corresponding to the main units.
The solar irradiation and PV array models give the power output from the
PV array, depending on location, tilt and azimuth angle, array type, and PV
modules. The inverter-pump model describes the variation in instantaneous
water flow of the pump, depending on input power from the PV modules and
13
working efficiency of the inverter and pump. The water demand model estimates the IWR and, together with the hydraulic head model, is crucial to design the pumping system and to evaluate daily hydraulic energy. Hydraulic
energy and solar irradiation allow for design of the PV array. The groundwater
model gives the response of groundwater resources to pumping in terms of
variation in water level. The crop growth model describes crop yield on the
basis of water supplied by the PVWP and irrigation systems and evaluates the
profitability of the system.
Dynamic simulations were conducted to quantify the matching between
IWR and water supply, to identify the effects of pumped water on groundwater
resources and crop yield, and to prove and optimize the design approach.
Figure 4:
Models for the design and simulation of PVWP systems for
irrigation.
3.1.1 Design of PVWP and WPWP systems
PVWP systems have been compared with WPWP systems since they are the
most widely used off-grid renewable pumping systems [56].
The design of PVWP and WPWP systems is a based on a detailed assessment of the IWR, and solar and wind energy resources. Based on the results
of previous works [30, 33], we improved the design approach for PVWP and
WPWP systems for irrigation. The proposed approach correlates PVWP and
WPWP capacity as a function of solar irradiation and wind speed.
14
The PVWP power peak Pp,PVWP (kWp) and the WPWP rated power Pr,WPWP
(kWr) can be calculated based on the following three equations:
𝑃𝑃𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 =
𝑃𝑃𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟
(𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼
𝐼𝐼𝐼𝐼𝐼𝐼𝑡𝑡𝑡𝑡𝑡
0.0027 𝑇𝑇𝑇𝑇𝑇𝑇
max
𝐸𝐸𝑠𝑠𝑚𝑚
𝑓𝑓𝑚𝑚 [1 − 𝛼𝛼𝑐𝑐 (𝑇𝑇𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 − 𝑇𝑇𝑆𝑆𝑆𝑆𝑆𝑆 )]𝜂𝜂𝑝𝑝 𝑚𝑚
(𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼
𝐼𝐼𝐼𝐼𝐼𝐼𝑡𝑡𝑡𝑡𝑡
0.0027 𝑇𝑇𝑇𝑇𝑇𝑇
=
𝑚𝑚𝑚𝑚𝑚𝑚
𝑚𝑚
𝐸𝐸𝑤𝑤𝑚𝑚
𝑓𝑓𝑚𝑚 𝜂𝜂𝑝𝑝
(𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼
𝐼𝐼𝐼𝐼𝐼𝐼𝑡𝑡𝑡𝑡𝑡
𝑥𝑥
(𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼
= ∑ 𝐼𝐼𝐼𝐼𝐼𝐼𝑗𝑗𝑗𝑗𝑗
𝑗𝑗𝑗𝑗
𝑚𝑚 𝑚 {𝐽𝐽𝐽𝐽𝐽𝐽𝐽 𝐽𝐽𝐽𝐽𝐽𝐽𝐽 𝐽 𝐷𝐷𝐷𝐷𝐷𝐷}
(1)
(2)
(3)
where 0.0027 is a conversion factor that takes into account the density of water
ρ (1000 kg/m3), gravity acceleration g (9.8 m/s2), and the conversion between
Joule and kWh (1/(3.6*106)) to calculate daily hydraulic energy; fm is the
matching factor assumed to be equal to 0.9 (the matching factor can be adjusted to also consider power losses during the lifetime of the PV generator
and other derating factors, such as soiling or shading); αC is the temperature
coefficient of the PV power, equal to 0.45 %/°C; Tcell is the PV cell temperature (°C); TSTC is the temperature under standard test conditions (25°C); ƞp
is the efficiency of the pump (%); IWRt,m represents total monthly average
daily irrigation water requirement (m3/ha/day) given by the sum of the IWR
of the j-th crops with x equal to total number of irrigated crops; TDH is total
dynamic head that takes into account the contributions of groundwater
depth, maximum drawdown, operational head of the irrigation system and
hydraulic losses (m); Es is the monthly average daily solar irradiation hitting
the PV array (kWh/m2/day); Ew is the specific monthly average daily energy
yield of the wind turbine (kWh/kWr/day); the function max indicates that the
design of the PVWP or WPWP systems has to be conducted for month m
marked out by the highest ratio (design ratio) between monthly average
daily IWR and monthly average daily solar irradiation or wind energy.
For sustainability purposes, IWRt,m has to be lower than the daily sustainable pumped water volume Vs,d to avoid overexploitation of groundwater resources. For technical reasons, the maximum drawdown induced during
pumping conditions smax has to be lower than the depth of the pumping system
hp (measured from the static water level). Vs,d and smax can be easily estimated
by running pumping tests. The superscript IPCC emphasizes that an accurate
design of both PVWP and WPWP systems should take into consideration
climate change conditions and thus the IWR at the end of the system
lifetime, according to Intergovernmental Panel on Climate Change (IPCC)
scenarios [57]. The total dynamic head TDH is calculated with the following
equation:
15
𝑇𝑇𝑇𝑇𝑇𝑇 = 𝑆𝑆𝑆𝑆 + 𝑠𝑠𝑚𝑚𝑚𝑚𝑚𝑚 + 𝐻𝐻𝑖𝑖𝑖𝑖𝑖𝑖 + ∑ 𝜆𝜆 𝜆𝜆 𝜆𝜆 𝜆
(4)
where SH is the static head (m); Hirr is the required head to operate the irrigation system (m); λ and ξ are continuous and concentrated head losses (m),
respectively; Hirr has been taken from a previous work conducted within our
research project and focused on the design of sprinkler and micro-irrigation
systems for grassland and farmland conservation [58].
The energy generated by the wind turbine Ew at wind speed v has been calculated from the power curve through the Weibull-based approach described
in Mathew [33]. The models rely on the assumption that wind turbine power
curves can be divided in two performance regions, with the first between the
cut-in vi (m/s) and rated speed vr (m/s), and the second between the rated vr
and the cut-out speed vo (m/s). Ew is given by the following equation [33]:
𝑣𝑣𝑟𝑟
𝑣𝑣𝑜𝑜
𝐸𝐸𝑤𝑤 = 𝑡𝑡 𝑡 𝑡𝑡𝑣𝑣 𝑓𝑓(𝑣𝑣)𝑑𝑑𝑑𝑑 + 𝑡𝑡𝑡𝑡𝑟𝑟 ∫ 𝑓𝑓(𝑣𝑣)𝑑𝑑𝑑𝑑
𝑣𝑣𝑖𝑖
𝑣𝑣𝑟𝑟
(5)
where t is the time period, assumed to be equal to one day (24 hours); v is the
wind speed (m/s); Pv is the power produced by the wind turbine in the power
curve region between vi and vr (kW); f(v) is the probability density function of
wind speed; and Pr is the wind turbine rated power (kWr). Pv and f(v) are given
by:
𝑃𝑃𝑣𝑣 = 𝑃𝑃𝑟𝑟
𝑓𝑓(𝑣𝑣) =
𝑝𝑝
𝑣𝑣 𝑝𝑝 − 𝑣𝑣𝑖𝑖
𝑝𝑝
𝑝𝑝
𝑣𝑣𝑟𝑟 − 𝑣𝑣𝑖𝑖
𝑘𝑘 𝑣𝑣 𝑘𝑘𝑘𝑘 −(𝑣𝑣) 𝑘𝑘
( )
𝑒𝑒 𝑐𝑐
𝑐𝑐 𝑐𝑐
(6)
(7)
where p is the velocity-power proportionality, assumed to be equal to 3; k is
the Weibull shape factor; and c is the scale factor. vi, vr and vo have been taken
from a survey of small wind turbine power curves [59].
3.1.1.1 Assessment of solar and wind energy potential
China has relatively high solar irradiation resources that make it suitable for
the implementation of PVWP systems for irrigation. Annual average horizontal solar irradiation varies greatly from 3–6 kWh/m2/day, depending on location and climate [60]. China is also characterized by considerable wind energy
potential, with annual average wind speeds of up to 7 m/s in the best locations
(wind speed at 10 m above ground) [60].
16
To assess available solar irradiation and wind speed at specific sites, Meteonorm [61] and Open Energy Information [60] were used as a climatic database and spatial dataset, respectively. PVsyst [46] and our developed codes
were used to calculate solar irradiation hitting the PV array depending on orientation angle and array technology. The optimization of PV array orientation
angles was mostly carried out to maximize solar irradiation gathered during
the irrigation season, typically from April to September.
3.1.1.2 Assessment of irrigation water requirement
The assessment of IWR is complex since it is affected by numerous factors,
such as climatic parameters (solar radiation, wind speed, temperature, and humidity), crop characteristics (type, variety, and development stage), management, and environmental conditions. Typically, the computation of IWR can
be performed by sequentially calculating reference evapotranspiration, evapotranspiration under standard cultural conditions, and effective rainfall [62].
Reference evapotranspiration ETo (mm/day) was calculated with the FAO
Penman–Monteith equation [63]:
900
0.408 𝛥𝛥𝛥(𝑅𝑅𝑛𝑛 − 𝐺𝐺) + 𝛾𝛾 𝑇𝑇 + 273 𝑣𝑣2 (𝑒𝑒𝑠𝑠 − 𝑒𝑒𝑎𝑎 )
𝑎𝑎
𝐸𝐸𝐸𝐸𝑜𝑜 =
𝛥𝛥 𝛥𝛥𝛥𝛥(1 + 0.34 𝑣𝑣2 )
(8)
where Rn is the monthly average daily net radiation at the soil surface
(MJ/m2/day); G is the soil heat flux density (MJ/m2/day); Ta is the monthly
average daily air temperature (°C); Δ is the saturation slope of the vapour pressure curve at T (kPa/°C); γ is the psychometric constant (kPa/°C); es is the
saturation vapour pressure (kPa); ea is the monthly average daily actual vapour
pressure (kPa); and v2 is the monthly average daily wind speed at 2 m above
ground (m/s). The crop water requirement (CWR) is given by the difference
between evapotranspiration under standard culture conditions ETc (mm/day),
and effective precipitation Pe (mm/day):
𝐶𝐶𝐶𝐶𝐶𝐶𝐶 𝐸𝐸𝐸𝐸𝑐𝑐 − 𝑃𝑃𝑒𝑒 = 𝐸𝐸𝐸𝐸𝑜𝑜 𝐾𝐾𝑐𝑐 − 𝑃𝑃𝑒𝑒
(9)
where Kc is the cultural coefficient dependent on crop growth stage. The effective precipitation Pe was calculated as 80% of total precipitation [64]. The
IWR (mm/day) (1 mm/day corresponds to 10 m3/ha/day, and ha denoted the
unit hectare) was determined using the following equation:
𝐼𝐼𝐼𝐼𝐼𝐼𝐼
𝐶𝐶𝐶𝐶𝐶𝐶
ƞ𝑖𝑖𝑖𝑖𝑖𝑖 (1 − 𝐿𝐿𝐿𝐿)
(10)
17
where ƞirr is the efficiency of the irrigation system assumed to be equal to 80%
[26]; LR is the leaching requirement needed to remove residual salts from the
root zone, assumed to be equal to 18% [65]. Hourly simulations of the IWR
can be performed by adjusting Equations 8−10 to hourly time step [63].
3.1.2 Simulation of PVWP systems
3.1.2.1 Modelling of the photovoltaic array
The global solar radiation on the tilted surface Gg,t (kW/m2) depends on the
global horizontal radiation as well as the surface orientation of the module,
and it consists of three components, namely: beam radiation Gb,t (kW/m2), diffuse radiation Gd,t (kW/m2), and ground-reflected radiation Gr,t (kW/m2) [66]:
𝐺𝐺𝑔𝑔𝑔𝑔𝑔 = 𝐺𝐺𝑏𝑏𝑏𝑏𝑏 + 𝐺𝐺𝑑𝑑𝑑𝑑𝑑 + 𝐺𝐺𝑟𝑟𝑟𝑟𝑟
(11)
The beam component of the global tilted radiation can be calculated from the
global horizontal radiation using the following equation [66]:
𝐺𝐺𝑏𝑏𝑏𝑏𝑏 =
𝐺𝐺𝑔𝑔𝑔𝑔 − 𝐺𝐺𝑑𝑑𝑑𝑑
cos(𝜃𝜃)
cos(90° − 𝛼𝛼)
(12)
1 + cos(𝛽𝛽)
2
(13)
where Gg,h is the global horizontal radiation (kW/m2); Gd,h is the diffuse horizontal radiation (kW/m2); α is the solar altitude (°); and θ is the angle of incidence (°).The angle of incidence θ is function of the declination angle δ (°),
latitude φ (°), tilt angle β (°), solar azimuth angle γs (°), surface azimuth angle
γsu (°) and hour angle ωh (°), and it has been computed according to the procedure described in Duffie and Beckman [67]. The diffuse and ground-reflected components of the global tilted radiation are given by the following
relations [66]:
𝐺𝐺𝑑𝑑,𝑡𝑡 = 𝐺𝐺𝑑𝑑,ℎ
𝐺𝐺𝑟𝑟,𝑡𝑡 = 𝜌𝜌𝑔𝑔 𝐺𝐺𝑑𝑑,ℎ
1 − cos(𝛽𝛽)
2
(14)
where ρg is the ground reflectance (%). The hourly values of the global horizontal radiation and the diffuse horizontal radiation have been taken as input
for this solar radiation model. The power output from the PV array PPV (kW)
is calculated by [67]:
𝑃𝑃𝑃𝑃𝑃𝑃 = ƞ𝑃𝑃𝑃𝑃 𝐴𝐴𝑃𝑃𝑃𝑃 𝐺𝐺𝑔𝑔𝑔𝑔𝑔
18
(15)
where ƞPV is the efficiency of the PV module (%); and APV is the PV array area
(m2), which depends on the installed peak power of the module. ƞPV is given
by the following equation [67]:
ƞ𝑃𝑃𝑃𝑃 = ƞ𝑃𝑃𝑃𝑃,𝑆𝑆𝑆𝑆𝑆𝑆 [1 +
𝜇𝜇
ƞ𝑃𝑃𝑃𝑃,𝑆𝑆𝑆𝑆𝑆𝑆
(𝑇𝑇𝑎𝑎 − 𝑇𝑇𝑆𝑆𝑆𝑆𝑆𝑆 ) +
− ƞ𝑃𝑃𝑃𝑃,𝑆𝑆𝑆𝑆𝑆𝑆 )𝐺𝐺𝑔𝑔𝑔𝑔𝑔 ]
(𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁 20)
(1
800
ƞ𝑃𝑃𝑃𝑃,𝑆𝑆𝑆𝑆𝑆𝑆
𝜇𝜇
(16)
where ƞPV,STC is the efficiency of the PV module at standard test conditions
(STC) (%); μ is the temperature coefficient of the PV efficiency (1/°C); Ta is
the ambient temperature (°C); and NOCT is the nominal operating cell temperature (°C). The temperature coefficient of the output power μ can be approximated by [67]:
𝜇𝜇 ≈ ƞ𝑃𝑃𝑃𝑃,𝑆𝑆𝑆𝑆𝑆𝑆
𝜇𝜇𝑉𝑉𝑜𝑜𝑜𝑜
𝑉𝑉𝑚𝑚𝑝𝑝
(17)
where μVoc is the temperature coefficient of the open-circuit voltage (V/°C);
and Vmp is the voltage at the maximum power point (V). The PV power output
was determined using a code, which calculates the solar radiation hitting a
surface and simulates PV systems as well as flat-plate and concentrating solar
thermal collectors.
3.1.2.2 Modelling of the inverter-pump subsystem
The PVWP systems simulated in this work are referred to PVWP systems
equipped with variable-frequency inverters and submersible centrifugal
pumps. This configuration is one of the most reliable solutions available in the
market, especially for high-capacity PVWP systems.
The amount of pumped water by a PVWP system significantly depends on
the dynamic variability of the solar radiation, temperature, and performances
of the inverter and the pumping system. The solar radiation and temperature
primarily affect the power output of the solar array, whereas the supplied
power affects the efficiencies of the pump and inverter.
The pump efficiency curve (water flow versus power input for a given
head) was calculated from the standard characteristic pump curve (head versus
water flow) according to the following set of equations, which are derived
from the affinity laws and were compiled by Abella et al. [68]:
𝐻𝐻𝑜𝑜
𝑄𝑄𝑜𝑜 = 𝑄𝑄𝑟𝑟 √
𝐻𝐻𝑟𝑟
(18)
19
𝑃𝑃𝑝𝑝𝑝𝑝𝑝 = 𝜌𝜌𝜌𝜌𝐻𝐻𝑜𝑜
𝑃𝑃𝑃𝑃𝑃𝑃 =
𝑄𝑄𝑟𝑟 𝐻𝐻𝑜𝑜
√
ƞ𝑝𝑝 𝐻𝐻𝑟𝑟
𝑃𝑃𝑝𝑝𝑝𝑝𝑝
ƞ𝑚𝑚𝑚𝑚𝑚 ƞ𝑖𝑖𝑖𝑖𝑖𝑖
ƞ𝑚𝑚𝑚𝑚𝑚 = ƞ𝑚𝑚𝑚𝑚𝑚 𝑃𝑃𝑝𝑝𝑝𝑝𝑝 √
(19)
(20)
𝐻𝐻𝑟𝑟
𝐻𝐻𝑜𝑜
(21)
5
100
4
80
3
60
2
40
Pump
Inverter
1
0
0
0.2
0.4
0.6
0.8
1
1.2
1.4
20
1.6
Inverter efficiency (%)
Pump water flow (m³/hour)
where Qo is the operational water flow (m3/hour) corresponding to the operational hydraulic head Ho (m); Qr is the reference water flow (m3/hour) at the
reference hydraulic head Hr (m), and obtained from the standard characteristic
pump curve; Pp,o is the operational pump power (kW); ƞm,o is the efficiency of
the motor at the working conditions (%); ƞm,r is the efficiency of the motor at
the reference working conditions (%); and ƞinv is the efficiency of the inverter
(%).
The efficiency of the inverter was taken from an inverter database [46] and
was used to describe its dynamic performance in terms of power loss as a
function of the power input. Examples of the inverter and pump efficiency
curves used for the simulations of the PVWP systems are depicted in Figure 5.
0
Power input (kW)
Figure 5:
Dynamic performances of the centrifugal pump and
inverter.
3.1.2.3 Modelling of the aquifer response to pumping
Typically, groundwater transient modelling relies on the Theis equation,
which gives the unsteady distribution of the drawdown s at a radial distance r
20
and time t under the following assumptions: (a) homogeneous and isotropic
confined aquifer; (b) no source recharging the aquifer; (c) compressible aquifer; (d) water released instantaneously as the head is lowered; and (e) constant
pumping flow rate [69]. However, the assumption of constant pumping rate is
unrealistic for PVWP systems, thus making the Theis equation inappropriate
for modelling the drawdown in this case. Instead, the solution computed and
validated by Rasmussen et al. [70] for sinusoidal pumping rate in a confined
aquifer was used to simulate the drawdown in this study. The choice of this
solution was driven by the fact that, in sunny conditions, the amount of
pumped water by a PVWP system has a trend similar to the positive part of a
sinusoid.
Using the pumped water and the characteristics of the aquifer as the inputs,
the following equation can be used to calculate the drawdown s:
𝑠𝑠(𝑟𝑟, 𝑡𝑡) =
𝑖𝑖𝜔𝜔𝑝𝑝
𝑄𝑄0
𝐾𝐾0 (𝑟𝑟√
)
2𝜋𝜋𝜋𝜋
𝐷𝐷
(22)
where r is the distance from the pumping well, assumed equal to 1 m; t is the
time variable (1 h); T is the aquifer transmissivity (m2/hour); K0 is the zeroorder modified Bessel function; i is the unit imaginary number; ωp is the
pumping frequency (given by the ratio of 2π to pc, the pumping cycle (hour))
(Hz); and D is the hydraulic diffusivity (m2/hour). The transmissivity T is a
measure of how much groundwater flows towards the well per unit surface
area during the pumping. The diffusivity D is given by the ratio of transmissivity to storativity S, which is a measure of the volume of water released per
unit surface area and per unit decline in hydraulic head [69].
3.1.2.4 Modelling of the crop growth
To evaluate the benefits of PVWP systems and perform comprehensive system optimizations, it is important to simulate the crop yield response to water
supply. Two different approaches have been used to model the crop response
to water: the first relies on a linear relationship introduced by FAO between
crop yield and water supply to predict the reduction in crop productivity [63];
the second uses AquaCrop, a software developed by FAO that simulates the
productivity of most herbaceous crops [50].
The crop-water production function relates the relative yield reduction to
the relative reduction in evapotranspiration and is given by the following relation [63]:
(1 −
𝐸𝐸𝐸𝐸𝑐𝑐,𝑎𝑎𝑎𝑎𝑎𝑎
𝑌𝑌𝑎𝑎
) = 𝐾𝐾𝑦𝑦 (1 −
)
𝑌𝑌𝑚𝑚
𝐸𝐸𝐸𝐸𝑐𝑐
(23)
21
where Ya is the actual yield (tonne DM/ha/year) (DM stands for dry matter);
Ym is the maximum yield (tonne DM/ha/year); Ky is the yield response factor;
and ETc,adj is the actual evapotranspiration (mm/day). Ya represents the crop
yield reduction from the maximum yield due to a reduction in the water provided through irrigation. Ky simplifies the complex natural procedures that influence the effect of water deficit on the crop productivity and is equal to 1.1
for Alfalfa [63]. ETc,adj depends on the water supply available to the crop
(through irrigation and rainfall) and on the soil parameters, and it can be calculated from ETc through the following equation:
𝐸𝐸𝐸𝐸𝑐𝑐,𝑎𝑎𝑎𝑎𝑎𝑎 = 𝐾𝐾𝑠𝑠 𝐸𝐸𝐸𝐸𝑐𝑐
(24)
where Ks is the water stress coefficient. The procedure adopted to calculate
the actual evapotranspiration ETc,adj is thoroughly described in Allen et al.
[63].
Unlike the crop-water production function, AquaCrop divides the actual
evapotranspiration, on a daily basis, into crop evaporation (responsible for
crop growth), which is a function of the canopy cover, and soil evaporation
(non-productive component of the evapotranspiration). The biomass yield is
thus a function of the crop transpiration and the water productivity parameter.
The actual yield is then calculated as product of the biomass yield and the
harvest index.
Detailed descriptions of the AquaCrop models can be found in Steduto et
al. [71] and in Raes et al. [72]. Although the current version of AquaCrop does
not support forage crops like Alfalfa, it is possible to simulate the yield of
Alfalfa, excluding multiple harvesting, using the option for leafy vegetable
producing crops. The crop parameters used in the growth modelling were
taken from a previous similar work conducted for Uzbekistan [73] and are
summarized in Table 1.
Table 1:
AquaCrop parameters for Alfalfa yield
simulation.
Crop parameter
Plant density
Initial canopy cover (CC0)
Fertility stress level
Reference harvest index (HI0)
Value
75,000 plants per ha
90%
50%
40%
3.2 Economic assessment of PVWP systems
The ICC was estimated based on component prices, structure costs, and engineering and installation costs. All the component prices used correspond to
Chinese companies, and, in general, to the Chinese market [74, 75]. All the
22
prices are expressed in US$. The LCC is estimated using the following equation:
𝐿𝐿𝐿𝐿𝐿𝐿 = 𝐼𝐼𝐼𝐼𝐼𝐼 + 𝑃𝑃𝑃𝑃(𝑜𝑜𝑜𝑜𝑜𝑜𝑎𝑎𝑎𝑎𝑎𝑎 ) + 𝑃𝑃𝑃𝑃(𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑎𝑎𝑎𝑎𝑎𝑎 )
(25)
where PW(omrann) and PW(fuelann) are the present worth of the annual operation, maintenance and replacement costs omrann ($) and annual fuel costs
fuelann ($), respectively.
The potential revenues that can be generated through the operation of a
PVWP irrigation system come from: forage, surplus electricity generated during the non-irrigation season, which can be combined with the available incentives for renewable energy production in China, and reduction of CO2
emissions. It should be noted that the possibility of selling the surplus electricity generated and receiving the corresponding incentives strictly depend on
the availability of a connected electric grid and specific incentive schemes for
standalone systems.
The utilization of PVWP systems for grassland irrigation can reduce CO 2
emissions in three possible ways: use of PV modules instead of diesel to power
the irrigation system, surplus electricity production from a renewable energy
resource, and soil carbon sequestration due to grassland rehabilitation. The
CO2 emission offsets achieved in this manner can then be monetized and
traded.
In this study, the profitability of PVWP systems was analysed using economic indicators such as the NPV, IRR and PBP, which can be calculated from
the following equations [33]:
𝑁𝑁
𝑁𝑁𝑁𝑁𝑁𝑁(𝑖𝑖𝑖𝑖) = −𝐼𝐼𝐼𝐼𝐼𝐼 + ∑
𝑛𝑛𝑛𝑛
𝐶𝐶𝐶𝐶
(1 + 𝑖𝑖𝑖𝑖)𝑛𝑛
(1 + 𝐼𝐼𝐼𝐼𝐼𝐼)𝑁𝑁 − 1
(1 + 𝐼𝐼𝐼𝐼𝐼𝐼)𝑁𝑁 − 1
] = 𝐼𝐼𝐼𝐼𝐼𝐼 + 𝑜𝑜𝑜𝑜𝑜𝑜𝑎𝑎𝑎𝑎𝑎𝑎 [
]
𝑅𝑅𝑎𝑎𝑎𝑎𝑎𝑎 [
𝐼𝐼𝐼𝐼𝐼𝐼(1 + 𝐼𝐼𝐼𝐼𝐼𝐼)𝑁𝑁
𝐼𝐼𝐼𝐼𝐼𝐼(1 + 𝐼𝐼𝐼𝐼𝐼𝐼)𝑁𝑁
𝑃𝑃𝑃𝑃𝑃𝑃 𝑃𝑃
ln (1 −
𝑖𝑖𝑖𝑖 𝐼𝐼𝐼𝐼𝐼𝐼
)
𝑅𝑅𝑎𝑎𝑎𝑎𝑎𝑎 − 𝑜𝑜𝑜𝑜𝑜𝑜𝑎𝑎𝑎𝑎𝑎𝑎
ln(1 + 𝑖𝑖𝑖𝑖)
(26)
(27)
(28)
where CF is the cash flow registered from year 1 to year N (year), given by
the annual difference between revenues and expenses ($); ir is the interest rate
(%); and Rann is the annual revenues ($).
23
3.3 Optimization of PVWP systems
The optimization of the PVWP system size for irrigation was conducted using
the GA with one objective function and one reliability function. The optimization parameters that were used for the GA are shown in Table 2.
Table 2:
Genetic algorithm (GA) parameters.
Parameter
Population size
Algorithm
Crossover type
Crossover rate
Selector
Mutation rate
Number of generations
Value
200
NSGA 2
Uniform random
50%
Crowded tournament
5%
200
NSGA-II is a multi-objective optimization algorithm based on non-dominated
sorting. It is the algorithm available in Solve XL to solve multi objectives
optimization problems [76].
The objective function maximizes the annual profit Pann ($), which is given
by the balance between Rann ($) and the sum of the annualized initial capital
cost ICCann ($) and omrann ($). The objective function thus maximizes Rann,
and consequently the crop yield Ya (tonne DM/ha/year), and minimizes the
sum of ICCann and omrann, and consequently the PVWP system size. The reliability function sets the system failures to zero, ensuring 100% reliability and
sustainability of the PVWP system during the whole irrigation season. A
PVWP system failure f is defined to occur if the hourly drawdown sh (induced
by the pumping system during the irrigation season) goes below the level of
the pump hp (measured from the static water level) or if the daily volume of
pumped water Vp,d is larger than Vs,d.
The mathematical formulation of the proposed optimization approach is
given by the following two relationships:
𝑚𝑚𝑚𝑚𝑚𝑚(𝑅𝑅𝑎𝑎𝑎𝑎𝑎𝑎 − 𝐼𝐼𝐼𝐼𝐼𝐼𝑎𝑎𝑎𝑎𝑎𝑎 − 𝑜𝑜𝑜𝑜𝑜𝑜𝑎𝑎𝑎𝑎𝑎𝑎 )
(29)
∑ 𝑓𝑓 = 0 (𝑓𝑓 = {0,1} , 𝑓𝑓 = 1 𝑖𝑖𝑖𝑖 𝑠𝑠ℎ > ℎ𝑝𝑝 𝑜𝑜𝑜𝑜 𝑉𝑉𝑝𝑝,𝑑𝑑 > 𝑉𝑉𝑠𝑠,𝑑𝑑 )
(30)
𝑅𝑅𝑎𝑎𝑎𝑎𝑎𝑎 = 𝑌𝑌𝑎𝑎 𝑝𝑝𝑓𝑓
(31)
Rann from forage sales depends on the actual forage yield Ya and the specific
forage price pf ($/tonne DM), according to the following equation:
24
ICC (1000 $)
Ya is a function of the pumped water, and thus the PVWP system capacity,
and was dynamically calculated, whereas the range of the specific forage price
pf was taken from a survey conducted in the US [55].
The ICC of the PVWP system and its components was estimated from the
PVWP system capacity, based on data provided by a manufacturing company
[74] and is depicted in Figure 6 as a function of the price of the PV modules
(1.0, 1.5, and 2.0 $/Wp). The project implementation costs were set equal to
30% of the cost of the PVWP components, including the cost of their design
and installation [74].
100
90
80
70
60
50
40
30
20
10
0
PVWP (PV module 1.0 $/Wp)
PVWP (PV module 1.5 $/Wp)
PVWP (PV module 2.0 $/Wp)
Inverter
Pump
0
5
10
15
20
25
30
PVWP capacity (kWp)
Figure 6:
Initial capital cost (ICC) of PVWP systems and PVWP
components as a function of the PVWP capacity.
The decision variables in the optimization of the PVWP system include: (a)
the PV peak power capacity, (b) the tilt angle β (°), and (c) the surface azimuth
angle γsu (°). The decision variables and the corresponding design spaces are
shown in Table 3.
Table 3:
Decision variables and corresponding
design spaces.
Decision variable
PV peak power capacity
Tilt angle β
Surface azimuth angle γsu
Boundaries
0–3.2 kWp
0°–50°
-30°–30°,
The PV peak power capacity has a range of 0–3.2 kWp, which directly affects
the ICC, volume of pumped water, and crop yield, and thus indirectly the annual revenues. The angles β and γsu are constrained to vary between 0° to 50°
and -30° to 30°, respectively, and they directly affect the harvested solar irradiation, and thus indirectly the PVWP system size.
25
3.4 Suitable and optimal locations for PVWP systems
A previous attempt to identify the technically suitable locations for implementing PVWP systems in China was conducted by the ADB [23, 77]. In this
“ADB approach”, the suitability was evaluated through spatial analysis of land
cover, terrain slope, rainfall, temperature, and sunshine hours. According to
the recommendations of this previous study [23, 77], the terrain slope of a
suitable location should be lower than 2–5% for furrow irrigation and lower
than 5–9% for micro-spray irrigation to avoid runoff, soil erosion, and water
and energy wastage. Additionally, the suitable annual precipitation for PVWP
irrigation systems was determined to be between 300 to 600 mm on the basis
of the grassland IWR. Moreover, the study stipulated that the annual ambient
temperature and sunshine hours should be lower than 20°C and 1400 h, respectively. The constraint on the annual ambient temperature is related to the
effect of temperature on the evapotranspiration and thus on the grassland IWR,
whereas the constraint on the sunshine hours ensures that the solar irradiation
available is sufficient to drive the PVWP system.
In this doctoral thesis, the technical suitability of grassland areas for the
implementation of PVWP systems was conducted using a spatial analysis similar to the ADB approach but with some novelties. Because the new approach
was developed at the International Institute for Applied Systems Analysis
(IIASA), it will be referred to hereafter as the “IIASA approach”. The following spatial data were taken into account in this approach: grassland distribution, terrain slope, precipitation, potential evapotranspiration, and water stress
index. As in the ADB approach, the grassland distribution and terrain slope
were used to identify potential grassland areas for the installation of irrigation
systems. However, unlike in the ADB approach, the grassland areas where
irrigation is required were determined by considering the difference between
precipitation and potential evapotranspiration (a difference lower than zero
means irrigation is required). Moreover, the spatial distribution of the water
stress index, defined as the ratio of water usage to sustainable water available
[78], was analysed to disregard those areas marked by high water stress (water
stress index higher than 0.1) and only select the areas where the implementation of PVWP systems would not cause an overexploitation of the groundwater resources. Table 4 shows the differences between the ADB and IIASA approaches in terms of the spatial data used to identify the technically suitable
grassland areas for PVWP irrigation systems.
The assessment of the technically suitable grassland areas was only a starting point for identifying the optimal areas, which were evaluated using the
techno-economic model BeWhere [54]. For the specific purpose of determining the optimal areas for the implementation of PVWP irrigation systems, the
original version of BeWhere was employed to minimize the supply chain cost
for forage production in China, considering the feasible grassland areas as forage supply locations. The cost of the entire forage production chain, including
26
the costs of PVWP system installation, feedstock management, transportation,
and additional forage to meet the forage demand, were taken into account in
this model. Moreover, the revenues from forage sale, sale of surplus electricity
generated by the PVWP system, and carbon trading were taken into consideration. In addition, the negative effects of CO2 emissions due to transportation
were also included in the model. The costs and CO2 emissions due to transportation were calculated considering the road and railroad networks [79], the
transportation distance and the specific transportation costs and emissions for
trucks and trains.
Table 4:
Spatial data used for the assessment of the technically suitable
grassland areas for PVWP systems for irrigation.
Definition
Grassland
Irrigation system
implementation
Required irrigation
area
Constraints
IIASA approach
Grassland distribution [79]
Terrain slope (<9˚) [80]
ADB approach
Grassland distribution [79]
Terrain slope ( <9˚) [80]
Precipitation – Potential evapo- Precipitation (300–600 mm) [80]
transpiration (<0) [80, 81]
Water stress index (<0.1) [78]
Temperature (<20˚C) [80]
Sunshine duration (>1400 h) [61]
The potential forage production was estimated from spatial data of the annual
net primary productivity (NPP) [82], using the approach described in Vadrevu
and Lasko [83]. To take into account the effect of irrigation on the NPP and
consequently on the Alfalfa yield, the experimental relationship, proposed by
Ozdogan [84], between the annual average irrigation and NPP increase, was
utilized. Moreover, the PVWP system capacity was assessed using the spatial
data of solar irradiation [60], reference evapotranspiration [85], precipitation
[80] and groundwater depth [86], which are all input data required for designing a PVWP system. The spatial ICC of the PVWP systems was then calculated from the spatial distribution of PVWP system capacity, taking into consideration the specific costs of the PVWP systems [74]. Finally, the spatial
distribution of PVWP system capacity was used to elaborate the design maps
for PVWP systems, which can be adopted as a basis for the optimal planning
of PVWP systems, providing a handy tool for policy makers, farmers, investors, and consultants interested in solar pumping applications.
A half-degree (equals 50×50 km at the Equator) spatial resolution grid was
used as the BeWhere input. Thus, the spatial resolution of the original map
was aggregated to a half-degree resolution. Only 1% of each cell grid was
considered as forage supply area in order to avoid high pressure on the available groundwater resources.
27
4 Results and discussions
This chapter presents the results of the field measurements carried out at a
pilot PVWP system in Inner Mongolia. It also discusses the effect of IWR on
the PVWP system design, the technical and economic feasibility of PVWP
systems in comparison with DWP and WPWP systems, the optimization of
PVWP systems, and the assessment of suitable and optimal locations for
PVWP systems in the Chinese grassland.
4.1 Field measurements
The PVWP system that was tested in this study is located in the Xilamuren
grassland area, Inner Mongolia, China (41.32° N, 111.22° E, and 1590 m
above mean sea level) and supplies water for the irrigation of a 1 ha field. The
main system components and characteristics, such as the peak power and orientation angles of the PV array, are listed in Table 5.
Table 5:
PVWP system components
and characteristics.
Data PVWP system
PV power (kWp)
Pump (kW)
Tilt angle (°)
Azimuth (°)
Value
1.44
1.1 (AC centrifugal)
42
-36
The PVWP system was tested under two different scenarios: the recirculation
scenario (S1), in which the water lifted by the pumping system was recirculated into the well, and the micro-irrigation scenario (S2), in which the water
lifted from the well was pumped directly into a micro-irrigation system located about 150 m away from the well. The aim of scenario S1 was to evaluate
the pumping system and to analyse the reliability of the models related to the
PVWP system, whereas the purpose of scenario S2 was to analyse the effects
of pumping on the groundwater level. To measure the pumped water from the
well, two flowmeters were installed along the pipeline network. Figure 7
shows a schematic diagram of the PVWP system test scenarios along with the
instruments used for gathering the operational data.
28
All the measurements performed and the corresponding instruments used
are listed in Table 6. The measured data for solar radiation, power output, and
water flow were used to validate the PVWP system model, particularly the
pump efficiency curve, whereas the measurements of the well water level were
used to evaluate the reliability of the model of groundwater level response to
pumping. In addition, the weighing lysimeter data was compared with the
modelled data of evapotranspiration to verify the model used to calculate the
IWR.
Figure 7:
Schematic diagram of the conducted field measurements.
Table 6:
Measurements carried out during the tests and the corresponding instruments, resolutions, and accuracies.
Measurements
Solar radiation
Power
Water flow
Well water table
Evapotranspiration
Instrument
Pyranometer
Power meter
Flowmeter
Pressure sensor
Lysimeter
Logging time
1 hour
1 hour
1 hour
1 hour
1 hour
Resolution
1 W/m2
1W
0.0001 m3/hour
0.02 mwc
0.01 mm
Accuracy
± 1%
± 1%
± 0.8%
± 1%
± 1%
Figure 8 compares the simulated and measured hourly water flow Qo vs. power
input data, which shows good agreement at power inputs lower than 800 W.
At higher PV power inputs, more deviation between the modelled data and
field data is observed. This can be attributed to the direct connection of the
29
pumping unit to the irrigation system, which can cause significant variation in
the operative hydraulic head at high flow rates. Percentage root mean square
error (PRMSE) of 11.1% was obtained between simulated and measured values using 250 measurements.
Water flow (m³/hour)
6
Modelled
5
Measured
4
3
2
1
0
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Power input (kW)
Reference evapotranspiration
(mm/hour)
Figure 8:
Measured and modelled hourly water flow vs. power input.
0.4
Modelled
Measured
0.3
0.2
0.1
0
0
10
20
30
40
50
60
70
Time (hour)
Figure 9:
Measured and modelled hourly reference evapotranspiration.
Figure 9 compares part of the simulated results and measured data for the reference evapotranspiration ETo. The calculated results agree well with the
measured data, and the discrepancies can be explained by factors such as wind
30
gusts, pressure on the lysimeter, objects on the lysimeter, and temperature effects. PRMSE of 4.2% was obtained between simulated and measured values
using 8760 measurements.
Figure 10 shows the modelled results and measured data for the drawdown,
as a function of water flow Qo. The model is in disagreement with the measured data for several factors: poor borehole construction, inadequate instrumentation, field conditions instead of laboratory conditions, and short time
period of the measurements (only 60 measurements). It should also be noted
that the model employed in this study is currently the most accurate in representing the variation in groundwater level due to the operation of a PVWP
system, which depends on the semi-sinusoidal daily solar radiation pattern.
Moreover, the same model was adopted by Rasmussen et al. [70] to simulate
the drawdown in sinusoidal pumping tests at different aquifers, and it showed
excellent agreement with the field test data.
Drawdown (m)
3
Modelled
Measured
2
1
0
0
1
2
3
4
5
Water flow (m³/hour)
Figure 10: Measured and modelled hourly groundwater level.
4.2 Impacts of irrigation water requirement on system
design
The extreme variability of solar irradiation, wind speed, and IWR during the
year makes the design of both PVWP and WPWP systems a complex and integrated procedure. Figure 11 shows how the IWR for Alfalfa varies during
the irrigation season, for a specific case investigated in Paper IV: Hails in Inner Mongolia, China. Correspondingly, the design ratio (defined to identify
the design month) varies based on the combined effect of IWR and solar irradiation and shows that June is the design month.
However, Alfalfa is a water-demanding crop, so the possibility of using
low water consumption crops should also be taken into account. In particular,
31
more water-efficient crops should be used when the main purpose of implementing PVWP systems is to prevent desertification and to rehabilitate grassland.
IWR
Design ratio PVWP
IWR (m³/ha/day)
70
60
16
14
12
50
10
40
8
30
6
20
4
10
2
0
MAY
JUN
JUL
AUG
SEP
Design ratio
80
0
Month
Figure 11: Irrigation water requirement (IWR) and design ratio for
PVWP systems in Hails, Inner Mongolia, China.
Figure 12 shows the difference between Alfalfa and pasture grass in terms of
IWR for Gancha in Qinghai, China. It is evident that the IWR for pasture grass
is significantly lower than that for Alfalfa, thus reducing the PVWP system
size (by 45%) and corresponding ICC. It is important to note that the lifetime
of a PVWP system is around 20 to 30 years, as it is bounded by the lifetime
of the most durable component, namely the PV modules. Thus, it is necessary
to take into account in the design phase the effects of future foreseen climate
changes, which could affect the annual amount and distribution of precipitation as well as the radiation pattern and ambient temperature, the main factors
that influence the IWR.
Figure 13 shows an example of how the IWR could vary in the future, comparing the Alfalfa IWR in 2014 to IWR in 2050, with the latter estimated using
the IPCC (Intergovernmental Panel on Climate Change) climate projections
for scenario A2 [57, 61]. It is observed that the peak IWR in 2050 for scenario
A2 is 55% higher compared with that in 2014, requiring a 55% increase in the
PVWP system size. The increase of the corresponding electricity requirements
for pumping during the whole irrigation season is 50%. The previous figure
(Figure 13) highlights the importance of including the projected future trend
of IWR as a technical input for the optimal design of PVWP systems. However, a more crucial aspect that should be considered is that the projected increase in IWR represents a threat for the food, water, and energy security of
32
China, as well as those countries that will experience the more severe effects
of climate change.
80
IWR (m³/ha/day)
70
IWR Alfalfa
60
IWR Pasture Grass
50
40
30
20
10
0
MAY
JUN
JUL
AUG
SEP
Month
Figure 12: Irrigation water requirement (IWR) for Alfalfa and
pasture grass in Gancha, Qinghai, China.
160
IWR (m³/ha/day)
140
120
IWR Alfalfa 2014
IWR Alfalfa 2050 - Scenario A2
100
80
60
40
20
0
MAY
JUN
JUL
AUG
SEP
Month
Figure 13: Current and future (2050 A2 IPCC scenario) trend of
Alfalfa irrigation water requirement (IWR) in Gancha,
Qinghai, China.
Leclère et al. [87] and Vorosmarty et al. [78] investigated the effects of climate
change on the global agricultural systems and concluded that the irrigated areas could greatly increase with the consequence of huge investments in the
irrigation, energy, and water resource management sectors. Fischer et al. [88]
evaluated the change in IWR based on the IPCC scenario A2 and found that
33
the IWR in East Asia could increase by 28 to 122 Gm3 from 2000 to 2080,
depending on the mitigation of greenhouse gas emissions and uncertainties in
model projection. Thus, from a climate change point of view, PVWP systems
could play a strategic role in providing electricity for irrigation purposes, both
in off- and on-grid areas.
Hourly dynamic simulations of IWR and pumped water are essential for
the evaluation of PVWP system performances, especially for ensuring that the
water supply matches the IWR. The results for a location in Xining are shown
in Figure 14. It can be seen that the designed PVWP systems fulfil the peak
IWR in June (design month), but there is surplus pumped water during the
remaining months of the irrigation season. This surplus, which is a direct consequence of the monthly variations in IWR and the design strategy of selecting
the PVWP system capacity based on the highest ratio of IWR to available solar
irradiation, presents a further optimization issue that needs to be addressed in
order to maximize the profit.
Pumped water and IWR
(1000 m³/ha/month)
2
Fixed DC 2.4 kWp
Tracking DC 1.8 kWp
IWR
1.8
1.6
Fixed AC 2.1 kWp
Tracking AC 1.6 kWp
1.4
1.2
1
0.8
0.6
0.4
0.2
0
MAY
JUN
JUL
AUG
SEP
Month
Figure 14: Pumped water and irrigation water requirement (IWR)
for different PVWP systems in Xining, Qinghai, China.
Potential solutions include using the same system for multiple applications or
designing the system not just based on the worst case but also taking into consideration the effects of reduced pumped water on the crop productivity. In
fact, the design of a PVWP system for irrigation should take into account the
specific purpose: grassland rehabilitation or farmland conservation. In the former case, reaching high productivities is not necessary, and it is enough to
improve the soil cover to halt desertification processes without overexploiting
the water resources. In the latter case, enhancing grass or forage productivity
34
is important to maximize the profits of the farmers. In both cases, the simulation of crop yield is essential for assessing the performance of the system,
from both technical and economic points of view.
Figure 15 shows an example of the correlation between PVWP capacities,
pumped water volume, and crop yield (Alfalfa). For desertification control,
low PVWP system capacities could be used to achieve small increases in the
grass soil coverage without overexploiting the groundwater resources. The optimal system size would depend on the grassland yield required to halt degradation processes and would thus be influenced by environmental and ecological considerations. On the other hand, higher PVWP system capacities could
be used for farmland applications.
Pumped water volume
(1000 m³/ha/year)
8
7
5
3
2
8
7
Pumped water volume
Crop yield
6
4
9
Farmland conservation
6
5
4
Desertification control
3
2
1
0
1
0.5
0.75
1
1.25
1.5
1.75
0
Crop yield (tonne DM/ha/year)
9
PVWP capacity (kWp)
Figure 15: Crop yield (Alfalfa) and pumped water volume as a
function of the PVWP system capacity.
4.3 Optimization of integrated PVWP systems
The effectiveness of the new approach proposed for optimizing PVWP systems for irrigation was tested using the pilot PVWP system installed in Inner
Mongolia as a case study. As investigated in Paper III, the existing PVWP
system showed to be oversized causing an excessive decline of the groundwater level in the well and thus continuous shutdowns. Thus, the existing PVWP
system was optimized assuming a maximum hourly drawdown sh corresponding to the depth of the pump hp measured from the static water level. Table 7
summarizes the characteristic parameters (decision variables, operation fail-
35
ures, ICC, Alfalfa yield, and PBP) for the existing and optimized PVWP systems. The PBP was calculated considering an interest rate of 6.4% [89], and a
project lifetime of 25 years. The project depreciation was not considered.
Table 7:
Existing and optimized PVWP systems characteristic parameters.
Parameter
Pp,PVWP (kWp)
Pump capacity (kW)
β (°)
γsu (°)
ICC ($)
f (times)
Ya (tonne DM/ha/year)
PBP (year)
Existing system
1.44
1.1
42
-36
4800
350
2.7
> lifetime
(forage price 207 $/tonne DM)
Optimized system
0.96
1.1
10
8
3900
0
2.6
16
(forage price 207 $/tonne DM)
380
375
370
365
360
355
350
345
340
1.00
0.99
0.98
0.97
0.96
0.95
0.94
Annual profit
PVWP capacity
0
2
4
6
8
10
0.93
12
0.92
Optimmal PVWP capacity
(kWp)
Pann ($)
The constraint due to the groundwater level represents the main limiting factor
on the PVWP capacity, which was reduced from the existing 1.44 kWp to 0.96
kWp. Figure 16 shows the progress of the genetic algorithm in selecting the
optimal PVWP capacity to maximise the annual profit Pann. The optimal solution was found after few generations due to the non-complex optimization
problem.
Number of generations
Figure 16: Progress of the genetic algorithm (GA).
The tilt angle was optimized from 42° to 10°, allowing an increase in the solar
irradiation harvested during the irrigation season from April to the end of July.
The optimization of the tilt and azimuth angles enabled a 10% increase in the
36
PV power output during the irrigation season, compared with the existing orientation. Figure 17 shows the effect of tilt angle and azimuth angle on the Rann
and the resulting optimal point.
50
-30
-20
40
Optimal point
30
20
10
-10
0
10
20
30
Tilt angle (°)
830-840
820-830
810-820
800-810
790-800
780-790
770-780
760-770
750-760
740-750
730-740
720-730
0
Azimuth angle (°)
Figure 17: Effect of tilt angle and azimuth angle on
the annual revenues.
The ICC for the optimized system is $3900, corresponding to a reduction of
18.8% from that of the current PVWP system. This reduction in investment
cost is mainly due to the lower investment required for the PV modules and
inverter and the resulting reduction in project implementation costs. In addition, the optimization guarantees continuous operation during the whole irrigation season. Compared to the 350 failures that the existing system showed,
the optimized system never reaches the set maximum hourly drawdown sh.
Figure 18 shows the trend of well water level for the optimized system during the entire irrigation season. Despite the decreased PV array size of the
optimized system, the effect on the crop yield is insignificant because of the
continuous shutdowns in the current PVWP system. The resulting annual crop
yields at the end of the irrigation season are 2.7 and 2.6 tonne DM/ha/year for
the existing and optimized systems, respectively. The optimized PVWP system also has a shorter PBP, which, along with the lower ICC, makes it more
cost-efficient than the existing system. The groundwater level constraint represents the determining factor influencing the optimal size of the PVWP system, assuming no system failures occur during the irrigation season. The other
sensitive parameters, such as the forage price and PV module price, can affect
the optimal PVWP system size but only if the drawdown is not a limiting factor. The groundwater level constraint narrows the search of the optimal PVWP
system capacity to a small range between 0.25 to 0.96 kWp. The crop yield
37
function, and thus the annual revenue, and the PVWP system cost function
show a linear trend in this range.
Well water level (m)
4
3.5
3
2.5
2
1.5
1
0.5
0
Pump depth
0
400
800
1200
1600
2000
2400
2800
Time (hour)
Figure 18: Well water level trend induced by the optimized system
during the irrigation season.
Accordingly, the variations in the forage and PV module prices have a negligible effect on the search for the optimal system size. However, if the groundwater level constraint could be disregarded, the variations in forage and PV
module prices can affect the optimal PVWP system capacity, as shown in Figure 19.
Optimal PVWP capacity (kWp)
Forage price ($/tonne DM)
2.16
100
150
200
250
300
350
400
2.16
2.14
2.14
2.12
2.12
2.1
2.1
2.08
2.08
2.06
2.06
2.04
2.02
2
0.75
2.04
Optimal PVWP capacity=f(PV module price)
2.02
Optimal PVWP capacity=f(Forage price)
1
1.25
1.5
1.75
2
2.25
2
PV module price ($/Wp)
Figure 19: Effect of forage price and PV module price on the optimal
PVWP systems size assuming no groundwater response
constraint.
38
The increase of the PV module price shifts the optimal point towards lower
capacities, whereas the increase of the forage price results in higher optimal
capacities. The sensitivity analysis results show the effectiveness and importance of the optimization approach, which also takes into account the economic aspects of the PVWP systems for irrigation. The results are significant
especially for large applications where the optimized design could bring substantial economic benefits.
4.4 Technical and economic feasibility of PVWP
systems
4.4.1 Comparison of standalone water pumping technologies
The economics of PVWP irrigation systems depend on the system components
and configuration as well as the benefits produced. Typically, the technical
choices involved in the installation of PVWP systems are concerned about the
use of DC or AC pumps, fixed PV or solar tracking arrays, and battery or
water storage systems. Comparisons between the AC and DC pumping systems and between the fixed and two-axis tracking PV arrays were conducted
in this study. The battery and water storage systems were not taken into account in this analysis because of the following considerations: the excellent
matching between IWR and PVWP system water supply (load matching), the
high water volumes involved in irrigation, and consequently the high investment costs of batteries or water tanks. Storage systems only become crucial
when the irrigation electric load has to be deferred from the PV electricity
production, such as in the case of night irrigation.
For PVWP applications, DC pumps are more expensive than AC pumps,
but the DC/DC converters required by DC pumps are cheaper than DC/AC
inverters. In terms of the PV array configuration, the PVWP systems equipped
with sun trackers have lower capital costs for the PV modules and the related
power conditioning system because of the greater harvested solar irradiation.
However, the tracker can increase the initial cost of the supporting system and
thus the related replacement and maintenance costs. To meet the IWR for 1 ha
of Alfalfa against a water head of 40 m in Xining, Qinghai Province, China,
four PVWP systems were designed and validated through dynamic simulations. These included a 2.4 kWp fixed system equipped with a DC pump, a 2.1
kWp fixed system equipped with an AC pump, a 1.8 kWp tracking system
equipped with a DC pump, and a 1.6 kWp tracking system equipped with an
AC pump. The results of the economic investigation on these systems is presented in Figure 20.
39
7
ICC (1000 $)
6
5
Engineering and installation
AC pump
DC/AC inverter
Structure
DC pump
DC/DC converter
PV tracking system
PV modules
4
3
2
1
0
Fixed DC 2.4 kWp Fixed AC 2.1 kWp Tracking DC 1.8
kWp
PVWP systems
Tracking AC 1.6
kWp
Figure 20: Initial capital cost (ICC) of the studied PVWP systems
for providing water to 1 ha Alfalfa in Xining, Qinghai,
China.
The most cost-effective solution was found to be the 2.1 kWp fixed array
PVWP system equipped with an AC pump. Although DC pumps require lessexpensive power control units, their high investment costs make them less
competitive. Another advantage of AC pumps is their large availability in the
market, which is a significant factor, especially for remote off-grid systems
for which the availability of spare components is an important issue. The PV
array with a tracking system can reduce the area of the PV modules by 30%
compared with the fixed PV array, but this abatement cannot compensate for
the higher investment cost, and consequently higher operation and maintenance costs of the sun tracker.
PVWP systems were also compared with DWP and WPWP systems. A 3.4
kWp PVWP system and a 5.3 kW DWP system were designed to provide water
for 1 ha of Alfalfa against a water head of 60 m in Dulan, Qinghai Province,
China. Figure 21 presents the variations of both ICC and LCC for these two
systems over the years. This analysis of ICC and LCC neglects the additional
investments related to the well and irrigation system and focuses only on the
PVWP and DWP systems. The diesel price, interest rate, and the project lifetime were set equal to 1.1$/l, 6.4% [89], and 25 years, respectively. The project depreciation was not considered.
Despite the lower operation and maintenance costs and zero fuel costs, the
high prices of the PV modules and inverter, coupled with the low diesel prices,
made the PVWP systems unaffordable compared with the DWP systems in
the 1990s in terms of both ICC and LCC. Although the price of solar cells
dropped and that of diesel increased by the middle of the 2000s, the PVWP
systems still could not compete with the DWP systems, but the cost gaps became smaller. Recently, the gap between the investment costs of the PVWP
40
and DWP systems has become even smaller because of the further drop in the
prices of PV modules and components. Despite the DWP system still being
more advantageous in terms of ICC, the PVWP system has started to be more
competitive in terms of LCC. Because of the rapid increase in diesel prices,
the LCC of DWP systems could be up to twice that of PVWP systems.
7
PVWP-ICC
DWP-ICC
PVWP-LCC
DWP-LCC
PV module price
60
50
40
6
5
4
30
3
20
2
10
1
0
1998
2005
PV module price ($/Wp)
ICC and LCC (1000 $)
70
0
2013
Year
Figure 21: Variation of initial capital cost (ICC) and life cycle
cost (LCC) of PVWP and DWP systems between
1998 and 2013.
Figure 22 shows the breakeven point analysis between PVWP and DWP ystems taking into account the effect of fuel tax and different annual diesel price
escalation rate.
40
PVWP
DWP
DWP (+4%)
DWP (+8%)
DWP (-4%)
DWP (no tax 2013)
DWP (no tax 2015)
LCC (1000 $)
35
30
25
20
15
10
5
0
0
5
10
15
20
25
Time (year)
Figure 22: Breakeven point analysis between PVWP and DWP systems.
41
The fuel tax in 2013 and escalation rates show a small effect on breakeven
point between PVWP and DWP systems, which occurs between the 5th and 7th
year. The tax share of diesel price in 2013 was 26.3% [90]. Excluding the fuel
taxes in 2013 shows a minor effect on the LCC of DWP system. The simultaneous effect of oil price fall in 2015 and fuel tax [91] shifts the breakeven
points to the 10th year. Fuel escalation rates show a notable impact on the LCC.
In particular, an escalation rate of 8% for DWP system results in a LCC of
about three times higher compared to PVWP system. Assuming an escalation
rate of -4% for the diesel price, the LCC of DWP system is about 60% higher
than PVWP system.
For a defined IWR of 50 m3/ha/day and a hydraulic head of 40 m, Figure
23 compares the ICC during the design month for PVWP and WPWP systems
as a function of the solar irradiation and wind speed, respectively, for different
PV module and wind turbine (WT) specific costs.
Average daily wind speed for the design month (m/s)
25
2
4
5
6
7
8
9
10
PV (0.8 $/Wp)
PV (1.3 $/Wp)
PV (2.0 $/Wp)
WT (1.0 $/Wr)
WT (1.3 $/Wr)
WT (5.5 $/Wr)
IWR=50 m³/ha/day
TDH=40 m
20
ICC (1000 $)
3
15
25
20
15
10
10
5
5
0
2
3
4
5
6
7
8
9
Average daily solar irradiation for the design month
(kWh/m²)
10
0
Figure 23: Initial capital cost (ICC) of PVWP and WPWP
systems as a function of the solar irradiation and
wind speed for the design month, and specific
costs of PV modules and wind turbine (WT).
Assuming a monthly average daily wind speed ranging between 5 to 7 m/s, a
5.5 $/Wr WPWP system can compete in terms of ICC with a 2.0 $/Wp PVWP
system only if the monthly average daily solar irradiation is lower than 2.5 to
4 kWh/m2/day, respectively. Comparatively, assuming a monthly average
daily solar irradiation ranging between 5 to 7 kWh/m2/day, a 1 $/Wr WPWP
42
IWR and pumped water
(1000 m³/ha/10-days)
system can compete with 0.8 $/Wp PVWP technology only if the monthly average daily wind speed is higher than 6.5 to 8 m/s. Furthermore, a 1.3 $ W p
PVWP system shows a similar ICC as 1.0 and 1.3 $/Wr WPWP systems if the
monthly average daily solar irradiation and wind speed range between 5 to 7.5
kWh/m2 and 5 to 7.5 m/s, respectively. These results show that the economic
effectiveness of PVWP and WPWP systems is tightly bounded to the availability of solar irradiation and wind speed, respectively, and the specific costs
of the main components. Assuming the lowest PV module and wind turbine
specific costs, WPWP systems are uncompetitive compared with PVWP systems in terms of ICC. Nevertheless, based on the average specific costs for
both PVWP and WPWP systems, it can be concluded that the most cost-effective solution for irrigation is site specific.
However, even though WPWP systems may be economically competitive,
they do not provide the same technical performance as PVWP systems. Dynamic simulations of the IWR and pumped water from PVWP and WPWP
systems were conducted to analyse the matching between IWR and supply and
to assess the effect of discrepancies on crop productivity. Figure 24 shows the
results in a 10-days period for the designed 2.9 kWp PVWP and 2.6 kWr
WPWP systems in Hails, Inner Mongolia, China. It is observed that the
WPWP system fails to meet the IWR during the 5th, 6th, and 8th 10-days periods. The mismatching between IWR and water supply is even more significant
on a daily basis because of the extremely variability of wind speed compared
with solar irradiation, as depicted in Figure 25.
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
IWR
PVWP
WPWP
1
2
MAY
3
4
5
JUN
6
7
8
9
10 11 12 13 14
JUL
AUG
SEP
Time (10-days period)
Figure 24: Mismatching between irrigation water requirement
(IWR) and pumped water from PVWP and WPWP
systems on 10-days period.
43
IWR and pumped water
(m³/ha/day)
160
IWR
PVWP
WPWP
140
120
100
80
60
40
20
0
0
10
20
30
Time (day)
Figure 25: Daily dynamic simulations of the irrigation water
requirement (IWR) and pumped water by PVWP
and WPWP systems in May.
The technical suitability of PVWP systems was also verified by simulations
conducted with AquaCrop for assessing the crop yield response to pumped
water. Assuming the IWR is met on a daily basis and disregarding for sustainability issues the water surplus, irrigation through the WPWP and PVWP systems can improve the Alfalfa yield from 1 tonne DM/ha/year to 6.7 and 7
tonne DM/ha/year, respectively. These results highlight the importance of irrigation for improving the crop productivity. Moreover, they demonstrate how
the mismatching between IWR and pumped water from the WPWP system
leads to a reduction in the crop yield.
4.4.2 Economic profitability
An example of profitability assessment for a PVWP irrigation system is depicted in Figure 26 as a function of the increased forage yield and price. The
forage productivities used in the profitability analysis were the increased forage yields registered in different pilot PVWP systems for grassland and farmland conservation in China [23]. The forage prices were set at 207 and
370 $/tonne DM, corresponding to the lowest and highest Alfalfa prices in
China in 2011 [55]. The interest rate and the project lifetime were set equal to
6.4% [89], and 25 years, respectively. The project depreciation was not considered.
44
60
14
50
12
10
NPV (207$/tonneDM)
NPV (370$/tonneDM)
PBP (207$/tonneDM)
PBP (370$/tonneDM)
30
20
8
6
10
4
0
PBP (year)
NPV (1000 $)
40
2
1.2
-10
3.8
8
12
0
Increased Alfalfa yield (tonne DM/ha/year)
Figure 26: Net present value (NPV) and payback period
(PBP) analysis.
Both the NPV and PBP are strictly dependent on the increased forage yield
and price. Considering the Alfalfa price at 207 $/tonne DM, the economic
profitability of the PVWP system is guaranteed only when the increase of Alfalfa yield is higher than 3.8 tonne DM/ha/year, with PBP varying between 13
and 2.5 years. If Alfalfa price is 370 $/tonne DM and the Alfalfa increased
yield is 3.8 tonne DM/ha/year, the NPV can be increased from $1600 to
$8160, and the PBP can be shortened from 13 years to 5.2 years. Taking into
account the best case of Alfalfa increased yield, 12 tonne DM/ha/year, the
PBP can vary between 2.5 years and 1 year depending on the Alfalfa market
price. Figure 27 shows the effect of the interest rate on the NPV.
3.8 tonne DM/ha/year-207 $/tonne DM
12.0 tonne DM/ha/year-370 $/tonne DM
NPV (1000 $)
100
90
80
70
60
50
40
30
20
10
0
-10
IRR=11%
0
20
IRR=84%
40
60
80
100
Interest rate (%)
Figure 27: Effect of the interest rate, and forage yield and
price on the net present value (NPV).
45
Considering the forage yield and price at 3.8 tonne/DM and 207 $/tonne DM,
respectively, the IRR is 11%. When Alfalfa yield and price are assumed equal
to 12.0 tonne/DM and 370 $/tonne DM, respectively, the IRR is 84%. In both
cases, PVWP system shows to be a profitable investment with the IRR higher
than the assumed interest rate.
Figure 28 shows the combined effect of the price of forage and PV modules
on the profitability of PVWP systems for the case investigated in Paper IV.
The crop yield was simulated with AquaCrop on the basis of the water supplied by the PVWP system. The assumed interest rate was 6.4% [89]. The
forage price has a notable effect on the system profitability only when the
prices of the PV modules are high. Thus, the lower the price of the PV modules, the lower the effect of forage price on the PBP. The 0.8 $/Wp PVWP
system has low PBP varying between 6 and 2 years showing high economic
profitability.
16
14
PV (0.8 $/Wp)
PV (1.3 $/Wp)
PV (2.0 $/Wp)
PBP (year)
12
10
8
6
4
2
0
200
250
300
350
400
Forage price ($/tonne DM)
Figure 28: Payback period (PBP) analysis as a function of the
forage price and specific costs of PV modules.
The results shown in Figures 26, 27 and 28 consider only the increased forage
productivity as revenue but at the same time disregard the ICC for the well
construction and irrigation system. It should be noted that the profitability of
PVWP systems for irrigation is very sensitive both to the other potential revenues generated during its operation (sale of surplus electricity, available incentives for renewable energy production, and carbon credits) and to the components included in the ICC (PVWP unit, well, and irrigation system).
The opportunity to harness the surplus electricity for multiple applications,
such as injection into the electric grid, and the benefit from the incentives for
renewable power generation can guarantee higher system profitability and
thus shorter the PBP. Although PVWP systems have been typically used for
46
NPV (1000 $)
off-grid applications, the recent drop in the price of PV modules could allow
the implementation of PVWP systems for on-grid applications as well, to meet
the energy requirements of the pastoral and agricultural sectors. Moreover, it
is worth to note that water pumping for irrigation is a controlled load. Thus, it
represents an interesting application that can largely increase the share of PV
electricity self-consumption, leading to shorter system PBP. Considering the
CO2 sequestration from grassland restoration and the CO2 emission reduction
from renewable electricity generation, the PVWP system investigated in Paper II could achieve CO2 emission savings of 7.4 tonne/ha/year. The corresponding offset could increase the profitability of the PVWP system depending on the carbon price. Figure 29 shows the effect of the interest rate on the
NPV, taking into account the ICC of PVWP system, well construction and
irrigation systems, as well as all the potential revenues. Thus, it provides a
comprehensive overview of the profitability of PVWP systems for grassland
and farmland irrigation. The results shown in Figure 29 refer to the cases and
scenarios investigated in Paper II.
120
110
100
90
80
70
60
50
40
30
20
10
0
-10
PVWP+W+I (3.8 tonne DM/ha/year-207 $/tonne DM)
PVWP+W+I (3.8 tonne DM/ha/year-207 $/tonne DM)/E-R-C
PVWP+W+I (12.0 tonne DM/ha/year-370 $/tonne DM)
PVWP+W+I (12.0 tonne DM/ha/year-370 $/tonne DM)/E-R-C
0
5
10
15
20
25
30
35
40
45
50
55
Interest rate (%)
Figure 29: Effect of the interest rate, forage yield and price, PVWP
systems components, and additional revenues on the net
present value (NPV) (W stands for well construction, I
for irrigation system, E for surplus of electricity generation, R for renewable electricity generation incentives, C
for carbon credits).
Assuming an increased forage productivity of 3.8 tonne DM/ha/year and a
forage price of 207 $/tonne DM, the investment which includes PVWP system, well and irrigation system is unprofitable independently from the interest
47
rate. The additional revenues due to surplus of electricity generation, corresponding incentives and carbon credits increase the profitability of the entire
system marked out by an IRR of 11%. If the increased forage productivity is
12.0 tonne DM/ha/year and the forage price is 370 $/tonne DM, the entire
PVWP irrigation system shows excellent profitability with and without considering the additional revenues, with IRR up to 54%.
4.5 Suitable and optimal locations for PVWP systems
The comparison between the technically suitable areas estimated using the
IIASA and the ADB approach is shown in Figure 30 in the form of a disagreement map, with the common areas highlighted in red. The total technically
suitable area for implementation of PVWP irrigation systems is almost similar
from both the approaches, about 25% of the total grassland areas. Nevertheless, there is a significant disagreement in the spatial distribution, especially
in the central provinces of China.
Legend
ADB
IIASA
Common areas
Figure 30: Disagreement map of the suitable areas for PVWP systems
identified by the IIASA and ADB approach.
According to the ADB approach, the technically suitable areas for grassland
irrigation in China are mainly located in the zone from the Northeast to the
Southwest of China. On the contrary, the technically suitable areas from the
48
IIASA approach are located in the Northeast of China, particularly in the
Northeast of Inner Mongolia, as well as in the Southwest of China, mainly in
Qinghai, Sichuan, and Tibet. This significant difference is linked to the fact
that the ADB approach neglects the availability of water resources in the assessment of suitable grassland areas. The calculated potential forage production from these suitable areas is 4.1 Mtonne DM, which corresponds to about
1% of the total livestock fodder demand in China and is about forty times
higher than China’s 2013 Alfalfa import volume of 100 ktonne DM [92]. The
results show that grassland irrigation, which is a crucial issue for the sustainable future of China and for enhancing rural economies, could increase the
internal forage supply to meet the demand for livestock or at least substantially
reduce the forage imports. Figure 31 shows the results of the optimization
process carried out using the BeWhere model in terms of forage production
required to meet the demand as a function of the forage market price.
Forage yield (Mtonne DM)
4.5
4
3.5
3
2.5
2
Carbon trading (100 $/tonne CO2)
50% ICC subsidy
Electricity + incentives
BAU
1.5
1
0.5
0
0
200
400
600
800
1000
1200
Forage price ($/tonne DM)
Figure 31: PVWP forage production as a function of the market
forage price for different studied scenarios.
Four different scenarios were considered: 1) business as usual (BAU) (forage
sale is the only income from the PVWP system operation); 2) 50% ICC (a
subsidy for 50% of the ICC is taken into account); 3) electricity + incentives
(the surplus electricity and the corresponding incentives for renewable power
production in China are considered as part of the revenues); and 4) carbon
trading (the CO2 emissions reduction and sequestration are considered as offsets at 100 $/tonne CO2).
BeWhere starts to select the optimal locations, and thus calculates the corresponding potential PVWP forage production, when the PVWP forage cost
is competitive with the market forage price. If the market forage price in-
49
creases, the number of locations selected as well as the PVWP forage production increases. Most of the locations are selected when the forage price ranges
from 150 to 300 $/tonne DM. The results for the first scenario indicated that
the forage price from PVWP irrigation is already competitive with the import
price, which varied from 200 to 400 $/tonne DM between 2009 and 2013, with
the US being the major exporter [55]. It is important to note that the import
prices are likely to increase further because of the high water requirements of
Alfalfa and the drought periods that have affected the US during the past years
[93]. The scenario with 50% ICC reduction shows only a slight effect on the
optimization process. This can be explained by the fact that the majority of the
locations selected are characterized by low IWR and shallow groundwater
level, and thus a smaller PVWP irrigation capacity. Similarly, the surplus electricity sales and incentives in the third scenario play only a marginal role in
the search for the optimal locations, which is tightly bounded to the PVWP
system capacity. Taking the offsets generated by carbon trading into consideration in the last scenario, the competitiveness of PVWP systems for forage
irrigation could be enhanced but only at high CO2 prices.
The identification of the optimal locations, and thus the corresponding forage production, is also sensitive to the potential NPP and, in particular, to the
potential improvement of NPP through irrigation. Figure 32 shows the localization of the optimal sites for the forage market prices of 200 and 400 $/tonne
DM.
Figure 32: Selection of the optimal locations for forage market price
200 (left) and 400 (right) $/tonne DM.
Most of the optimal locations for the implementation of PVWP technology
are mainly situated in Tibet, Qinghai and Sichuan Provinces. Locations in the
Northeast of China are selected but only at high forage market price.
50
4.5.1 PVWP system design maps
The PVWP system design maps have been developed as a rapid tool for decision makers, investors, farmers, and consultants interested in PVWP applications. These maps provide spatial information about the PVWP system capacity, and thus implicitly about the corresponding ICC, based on the spatial distribution of the following data: solar irradiation, reference evapotranspiration,
precipitation, and groundwater depth.
The PVWP system design was conducted using the approach described in
Section 3.1.1. The peak IWR was calculated from ETo assuming June as the
design month, Alfalfa as the reference crop, and the effective precipitation as
80% of the total precipitation. To estimate the TDH, smax and Hirr were assumed equal to 10 m and 20 m [58], respectively. The resulting PVWP system
design map for irrigating Alfalfa in Qinghai Province is shown in Figure 33.
Figure 33: PVWP system design map for the irrigation of 1 ha Alfalfa
in Qinghai Province, China.
51
The PVWP system capacity refers to the irrigation of 1 ha. The map provides
quick information on the spatial distribution of the PVWP system capacity for
irrigating Alfalfa, with a resolution of 1 km2. Similarly, PVWP system design
maps can be created for different crops or for several crop rotations. By including the spatial information of the potential NPP and potential increase of
NPP due to irrigation, the map can easily show the spatial profitability of implementing PVWP system for irrigation. To improve the accuracy of PVWP
system design maps, the projected future trend of the crop IWR, the ground
water quality and availability, and the probable future decline of the groundwater should also be taken into account to avoid undersized PVWP systems.
52
5 Summary of papers
This section summarizes the main results of each paper appended to this doctoral thesis, highlighting the author’s and co-authors’ contributions. I served
as the first author of all appended papers performing most of the writing. My
main supervisor Prof. Jinyue Yan and co-supervisor Assoc. Prof. Hailong Li
reviewed all the appended papers together with all my co-authors.
5.1 Paper I
P.E. Campana, H. Li, J. Yan, “Dynamic modelling of a PV pumping system
with special consideration on water demand”, Applied Energy, vol. 112, p.
635–645, 2013.
5.1.1 Division of the work
I was the lead author of the paper. I proposed the topic of the paper. I built the
model for the dynamic simulations of the PVWP system and of the IWR. I
conducted all the simulations and the economic analysis.
5.1.2 Results and discussions
Paper I validates a simplified approach for designing PVWP systems using
dynamic modelling. The technical effectiveness of the designed PVWP systems was evaluated in terms of matching between water supply and IWR. The
models of PVWP systems and IWR were thus presented. To identify the best
PVWP systems technical configuration, fixed PV array was compared to two
axis tracking PV array as well as DC pump was compared to AC pump. The
conducted simulations and economic analysis showed that fixed PV array and
AC represent the most cost effective and reliable components for remote
PVWP systems.
The paper highlights the issue that the common PVWP systems design approach results in a surplus of the water supply both during the irrigation season
and non-irrigation season. The highlighted issues opened the discussion on
how to harness the surplus of electricity to further improve the economic competitiveness of PVWP systems.
53
5.2 Paper II
P.E. Campana, A. Olsson, H. Li, J. Yan, “An economic analysis of photovoltaic water pumping irrigation systems”, International Journal of Green Energy, 2015 (Article in press).
5.2.1 Division of the work
I was the lead author of the paper. I proposed the topic of the paper. I gathered
all the economic data and I conducted all the economic analyses. Alexander
Olsson from the Royal Institute of Technology (KTH) contributed in estimating the amount of CO2 sequestered by the improved grassland productivity
using PVWP systems for irrigation.
5.2.2 Results and discussions
Paper II compares PVWP systems to DWP systems from the perspectives of
the investment cost, life cycle cost and profitability. The paper provides a
comprehensive analysis of the possible revenues generated by the operation
of PVWP systems for grassland irrigation and how they affect the profitability.
The environmental benefits of using PVWP for degraded grassland desertification were evaluated and monetized. Results showed that, although the
PVWP systems for irrigation can have low payback periods, their profitability
is affected by several factors such as: forage yield and price, system components included in the initial capital cost, subsidies for the surplus of electricity
generation and CO2 credits.
5.3 Paper III
P.E. Campana, H. Li, J. Zhang, R. Zhang, J. Liu, J. Yan, “Economic optimization of photovoltaic water pumping systems for irrigation”, Energy Conversion and Management, vol. 95, p. 32–41, 2015.
5.3.1 Division of the work
I was the lead author of the paper. I proposed the topic of the paper. I built the
optimization model of the PVWP system for irrigation. I gathered all the technical and economic data. I performed all the field measurements with the help
of Jun Zhang from the Chinese Institute of Water Resources and Hydropower
Research (IWRHR). Prof. Zhang from the Institute of Water Resources for
Pastoral Areas (IWRPA) in Inner Mongolia provided all the facilities and
54
technical equipment for running the field measurements. Prof. Liu from
IWRHR supervised the study.
5.3.2 Results and discussions
Paper III proposes a new integrated optimization approach for the design of
PVWP systems with special consideration of groundwater constraint and crop
yield under different water supply. An hourly optimization model based on
GA was developed. The field tests conducted at a pilot PVWP system in Inner
Mongolia were used to analyse the suitability of the dynamic models adopted
to simulate the PVWP system operation. Results showed that the groundwater
level constraint determines the optimal size of the PVWP system. The PV
modules and forage price can affect the optimal PVWP system size but only
if the groundwater is not a limiting factor.
5.4 Paper IV
P.E. Campana, H. Li, J. Yan, “Techno-economic feasibility of the irrigation
system for the grassland and farmland conservation in China: photovoltaic vs.
wind power water pumping”, manuscript for journal consideration.
5.4.1 Division of the work
I was the lead author of the paper. I proposed the topic of the paper. I built the
model for the design of PVWP and WPWP systems. I conducted all the simulations of the PVWP and WPWP systems, IWR and crop yield response. I
conducted all the economic analysis.
5.4.2 Results and discussions
Paper IV compares the technical effectiveness and economic feasibility of
PVWP and WPWP systems for irrigation of the degraded grassland in China.
A new general approach for the design and selection of the suitable technology
for irrigation purposes was developed calculating the PVWP and WPWP system sizes based on IWR, solar irradiation and wind speed. The paper combines
the PVWP and WPWP systems simulations with the crop growth simulations
to estimate the benefit of irrigation on the forage yield.
The results showed that, based on the current WPWP system components
costs, WPWP system can compete with PVWP system only at high wind
speed and low solar irradiation values. Dynamic simulations showed also that
PVWP system represent the best technical solution to provide water for irrigation due to the best matching between water supply and IWR that positively
affect the crop yield and thus revenues.
55
5.5 Paper V
P.E. Campana, S. Leduc, M. Kim, J. Liu, F. Kraxner, I. McCallum, H. Li, J.
Yan, “Optimal grassland locations for sustainable photovoltaic water pumping
systems in China”, Proceedings of the 7th International Conference on Applied
Energy (ICAE 2015), March 28–31 2015, Abu Dhabi, United Arab Emirates.
5.5.1 Division of the work
I was the lead author of the paper. I proposed the topic of the paper. I conducted all the spatial analyses with the help of Moonil Kim from Korea University. I gathered and set all the input parameters for BeWhere and I conducted all the optimizations. Sylvain Leduc from IIASA contributed in setting
the BeWhere model for the specific purpose of studying the forage supply
chain using PVWP systems as irrigation technology. Prof. Liu from Beijing
Forestry University contributed with suggestions on how to identify the most
technically suitable locations for implementing PVWP systems in the Chinese
grassland. Florian Kraxner and Ian McCallum from IIASA supervised the
study.
5.5.2
Results and discussions
Paper V proposes a new methodology to identify the suitable and optimal
grassland areas for implementing PVWP systems for irrigation using spatial
analysis. The most suitable locations were used as starting point for selecting
the optimal locations capable to minimize the forage supply chain costs in
China.
The results showed that there is a significant disagreement between our
novel approach (IIASA approach) and the previous approach (ADB approach)
in the spatial distribution of the suitable areas for implementing PVWP systems for irrigation. This significant difference was related to the exclusion of
the spatial distribution of water resources availability in the ADB approach.
The optimal location resulted very sensitive to several parameters such as forage price, CO2 credits, groundwater level, potential NPP and to the potential
improvement of NPP through irrigation.
56
6 Conclusions
The PVWP system is a renewable energy based solution to support the sustainable development of the pastoral and agricultural sectors in China and to
face the future foreseen negative effects of climate change on the IWR. The
PVWP system has the capability of preventing desertification while also generating profit to improve the incomes and thus living conditions of farmers,
especially in the remote areas.
The main conclusions of this doctoral thesis are the following:






An accurate assessment of the crop irrigation water requirement is essential for the optimal design of PVWP systems. Moreover, analysing
the effects of pumped water on the available water resources and crop
productivity is essential for verifying the technical successfulness of
the design.
The constraint related the groundwater resources represents the main
determining factor for the optimal size of PVWP systems.
As a result of the recent decrease in the PV module prices, PVWP
systems have become more competitive than the traditional off-grid
water pumping technologies.
The profitability analyses showed that PVWP systems can have high
positive NPV within the lifetime of the investment and thus low PBP
(down to 2 years). However, the profitability of a PVWP system is
sensitive to several factors such as the interest rate, forage price, system components included in the ICC, possibility of harnessing the
surplus electricity, incentives for renewable power production and
carbon credits.
The potential value of PVWP systems for grassland conservation and
increasing productivity are substantial for China, both in terms of forage production (up to 4 Mtonne) and in terms of CO2 emissions savings and sequestration.
The optimal locations for implementing PVWP systems are sensitive
to several environmental and economic parameters such as the forage
potential yield, forage water requirements, increased forage yield after irrigation, groundwater depth, forage price, and CO2 price. Therefore, these factors need to be carefully considered along with the selection of the most technically suitable areas.
57
7 Limitations and future works
This doctoral thesis represents a first attempt at studying the suitability of
PVWP technology for curbing the progress of grassland desertification and
supporting the sustainable development of farmland in China. This study thus
provides an overview on how to design, optimize, and assess the technical and
economic feasibility of PVWP systems. This section discusses some of the
limitations of this study and the issues that require further investigation.
The PVWP and WPWP systems were separately analysed and compared in
this study. However, the possibility of using hybrid PV-wind-storage systems
for irrigation and optimizing the two component systems based on cost and
reliability needs to be evaluated. More efforts are also required to study the
possibility of harnessing the entire power system for multiple applications,
such as providing power for irrigation as well as for other off-grid loads. This
could be economically beneficial, since irrigation is required only for a few
months of the year.
The environmental feasibility of PVWP systems has to be further analysed.
In particular, the effects of climate change on the IWR, and thus on the PVWP
system design and operation, need to be investigated in more detail. Moreover,
the spatial analysis conducted to identify the suitable areas for implementing
PVWP systems can be further improved by adding other influential factors,
such as the quality of groundwater resources and distribution of power lines,
which were not taken into account in the current work. In addition, the search
for the optimal locations for PVWP systems was limited to the grassland areas,
but it could be extended to the farmland in China.
Although this thesis concentrates on off-grid systems, there is an interesting
potential for interfacing and connecting PVWP systems to the grid. Thus, the
feasibility of integrating PVWP systems in on-grid areas needs to be analysed,
and control systems for PVWP technology need to be developed in future
studies. The aim of such a control system would be to enhance the technical
and economic effectiveness of the PVWP systems for irrigation while allowing the connection of the systems to the grid when irrigation is not required.
The control system could also be used for active management of the matching
between IWR and water supply and the groundwater response to pumping.
More effort is also required for the development of standalone monitoring systems to control and visualize the operational parameters of the system as well
as the effects of the system operation on the environment, particularly the effects of pumping on the water resources and crop yield.
58
References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
J. Li, X. Yang, Y. Jin, Z. Yang, W. Huang, L. Zhao, T. Gao, H. Yu, H. Ma, Z.
Qin, B. Xu, “Monitoring and analysis of grassland desertification dynamics
using Landsat images in Ningxia, China”, Remote Sensing of Environment,
vol. 138, p. 19–26, 2013.
T. Akiyama, K. Kawamura, “Grassland degradation in China: Methods of
monitoring, management and restoration”, Japanese Society of Grassland Science, vol. 53, p. 1–17, 2007.
X.S. Yi, G.S. Li, Y.Y. Yin, “The impacts of grassland vegetation degradation
on soil hydrological and ecological effects in the source region of the Yellow
River-A case study in Junmuchang region of Maqin country”, Procedia Environmental Sciences, vol. 13, p. 967–981, 2012.
C. Gang, J. Zhang, J. Li, “The advances in the carbon source/sink researches
of typical grassland ecosystem in China”, Procedia Environmental Sciences,
vol. 10, p. 1646–1653, 2011.
S. Li, P.H. Verburg, S. Lv, J. Wu, X. Li, “Spatial analysis of the driving factors
of grassland degradation under conditions of climate change and intensive use
in Inner Mongolia, China”, Regional Environmental Change, vol. 12, p. 461–
474, 2012.
D. Xu, C. Li, X. Song, H. Ren, “The dynamics of desertification in the farmingpastoral region of North China over the past 10 years and their relationship to
climate change and human activity”, Catena, vol. 123, p. 11–22, 2014.
J.M. Suttie, S.G. Reynolds, C. Batello, “Grasslands of the world”, FAO, 2005.
A.K. Knapp, J.M. Briggs, J.K. Koelliker, “Frequency and extent of water limitation to primary production in a mesic temperate grassland”, Ecosystems, vol.
4, p. 19–28, 2001.
R.J. Martin, S.M. Thomas, D.R. Stevens, R.F. Zyskowski, D.J. Moot, T.J. Fraser, “Improving water use efficiency on irrigated dairy farms in Canterbury”,
Proceedings of the New Zealand Grassland Association, 2006.
J.W. Heard, M.J. Porker, D.P. Armstrong, L. Finger, C.K.M. Ho, W.J. Wales,
B. Malcolm, “The economics of subsurface drip irrigation on perennial pastures and fodder production in Australia”, Agricultural Water Management,
vol. 111, p. 68–78, 2012.
S. Khan, M.A. Hanjra, J. Mu, “Water management and crop production for
food security in China: A review”, Agricultural Water Management, vol. 96,
p. 349–360, 2009.
R. Lal, “Food security in a changing climate”, Ecohydrology & Hydrobiology,
vol. 13, p. 8–21, 2013.
B. D. Vick, B.A. Neal, “Analysis of off-grid hybrid wind turbine/solar PV water pumping systems”, Solar Energy, vol. 86, p. 1197–1207, 2012.
R. Fostera, A. Cota, “Solar water pumping advances and comparative economics”, Energy Procedia, vol. 57, p. 1431–1436, 2014.
59
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
60
United Nations, “UNCCD-United Nations convention to combat desertification in those countries experiencing serious drought and/or desertification, particularly in Africa”, United Nations, 1997.
S. G. Li, Y. Harazono, T. Oikawaa, H. L. Zhao, Z. Y. He, X. L. Chang, “Grassland desertification by grazing and the resulting micrometeorological changes
in Inner Mongolia”, Agricultural and Forest Meteorology, vol. 102, p. 125–
137, 2000.
T.L. Bo, L.T. Fu, X.J. Zheng, “Modeling the impact of overgrazing on evolution process of grassland desertification”, Aeolian Research, vol. 9, p. 183–
189, 2013.
X. Wang, F. Chen, Z. Dong, “The relative role of climatic and human factors
in desertification in semiarid China”, Global Environmental Change, vol. 16,
p. 48–57, 2006.
C. Longjun, “UN Convention to Combat Desertification”, Encyclopedia of Environmental Health, p. 504–517, 2011.
Millennium Ecosystem Assessment, “Ecosystems and human well-being: Desertification synthesis”, World Resources Institute, Washington, DC, 2005.
S. Cao, L. Chen, D. Shankman, C. Wang, X. Wang, H. Zhang, “Excessive reliance on afforestation in China's arid and semi-arid regions: Lessons in ecological restoration”, Earth-Science Reviews, vol. 104, p. 240–245, 2011.
W. Mekuria, E. Veldkamp, M. Haile, J. Nyssen, B. Muys, K. Gebrehiwot, “Effectiveness of exclosures to restore degraded soils as a result of overgrazing in
Tigray, Ethiopia”, Journal of Arid Environments, vol. 69, p. 270–284, 2007.
J. Yan, H. Wang, Z. Gao, “Qinghai pasture conservation using solar photovoltaic (PV)-driven irrigation”, Asian Development Bank (ADB), 2010.
C. Gopal, M. Mohanraj, P. Chandramohan, P. Chandrasekar, “Renewable energy source water pumping systems-A literature review”, Renewable and Sustainable Energy Reviews, vol. 25, p. 351–370, 2013.
P.A. Lynn, “Electricity from sunlight”, Wiley, 2010.
L. Karlberg, F.W.T. Penning de Vries, “Exploring potentials and constraints
of low-cost drip irrigation with saline water in sub-Saharan Africa”, Physics
and Chemistry of the Earth, vol. 29, p. 1035–1042, 2004.
L.C. Kelley, E. Gilbertson, A. Sheikh, S.D. Eppinger, S. Dubowsky, “On the
feasibility of solar-powered irrigation”, Renewable and Sustainable Energy
Reviews, vol. 14, p. 2669–2682, 2010.
F. Cuadros, F. Lopez-Rodriguez, A. Marcos, J. Coello, “A procedure to size
solar-powered irrigation (photoirrigation) schemes”, Solar Energy, vol. 76, p.
465–473, 2004.
A. Hamidat, B. Benyoucef, T. Hartani, “Small-scale irrigation with photovoltaic water pumping system in Sahara regions”, Renewable Energy, vol. 28, p.
1081–1096, 2003.
Z. Glasnovic, J. Margeta, “A model for optimal sizing of photovoltaic irrigation water pumping systems”, Solar Energy, vol. 81, p. 904–916, 2007.
T. Meier, “Policy Recommendations to Improve the Sustainability of Rural
Water Supply Systems”, IEA, 2011.
I. Odeh, Y.G. Yohanis, B. Norton, “Economic viability of photovoltaic water
pumping systems”, Solar Energy, vol. 80, p. 850–860, 2006.
S. Mathew, “Wind Energy-Fundamentals, Resource Analysis and Economics”, Springer, 2006.
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
B. Bouzidi, “Viability of solar or wind for water pumping systems in the Algerian Sahara regions – case study Adrar”, Renewable and Sustainable Energy
Reviews, vol. 15, p. 4436–4442, 2011.
R. Diaz-Mendez, A. Rasheed, M Peillon, A. Perdigones, R. Sanchez, A.M.
Tarquis, J.L. Garcia-Fernandez, “Wind pumps for irrigating greenhouse crops:
Comparison in different socio-economical frameworks”, Biosystems Engineering, vol. 128, p. 21–28, 2014.
A. Mahjoubi, R.F. Mechlouch, B. Mahdhaoui, A.B. Brahim, “Real-time analytical model for predicting the cell temperature modules”, Sustainable Energy
Technologies and Assessments, vol. 6, p. 93–104, 2014.
S.A.K.H. Mozaffari Niapour, S. Danyali, M.B.B. Sharifian, M.R. Feyzi,
“Brushless DC motor drives supplied by PV power system based on Z-source
inverter and FL-IC MPPT controller”, Energy Conversion and Management,
vol. 52, p. 3043–3059, 2011.
J. Fernandez-Ramos, L. Narvarte-Fernandez, F. Poza-Saura, “Improvement of
photovoltaic pumping systems based on standard frequency converters by
means of programmable logic controllers”, Solar Energy, vol. 84, p. 101–109,
2010.
A. Kordzadeh, “The effects of nominal power of array and system head on the
operation of photovoltaic water pumping set with array surface covered by a
film of water”, Renewable Energy, vol. 35, p. 1098–1102, 2010.
S. Ould-Amrouche, D. Rekioua, A. Hamidat, “Modelling photovoltaic water
pumping systems and evaluation of their CO2 emissions mitigation potential”,
Applied Energy, vol. 87, p. 3451–3459, 2010.
H. Xu, J. Liu, D. Qin, X. Gao, J. Yan, “Feasibility analysis of solar irrigation
system for pastures conservation in a demonstration area in Inner Mongolia”,
Applied Energy, vol. 112, p. 697–702, 2013.
X. Gao, J. Liu, J. Zhang, J. Yan, S. Bao, H. Xua, T. Qin, “Feasibility evaluation
of solar photovoltaic pumping irrigation system based on analysis of dynamic
variation of groundwater table”, Applied Energy, vol. 105, p. 182–193, 2013.
T. Gao, R. Zhang, J. Zhang, “Effect of Irrigation on Vegetation Production and
Biodiversity on Grassland”, Procedia Engineering, vol. 28, p. 613 – 616, 2012.
C. Zhang, J. Yan, “Business model innovation on the photovoltaic water pumping systems for grassland and farmland conservation in China”, Energy Procedia, vol. 61, p. 1483 – 1486, 2014.
J. Yang, A. Olsson, J. Yan, B. Chen, “A Hybrid Life-Cycle Assessment of CO2
Emissions of a PV Water Pumping System in China”, Energy Procedia, vol.
61, p. 2871–2875, 2014.
PVsyst. Available at: http://www.pvsyst.com/en/. Accessed: 16/01/2015
Trnsys. Available at: http://www.trnsys.com/. Accessed: 16/01/2015
Matlab. Available at: http://www.mathworks.com/. Accessed: 16/01/2015.
Cropwat. Available at: http://www.fao.org/nr/water/infores_databases_cropwat.html. Accessed: 16/01/2015
AquaCrop. Available at: http://www.fao.org/nr/water/aquacrop.html. Accessed: 16/01/2015.
SolveXL. Available at: http://www.solvexl.com/. Accessed: 16/01/2015.
ArcGis. Available at: https://www.arcgis.com/features/. Accessed:
16/01/2015.
GAMS. Available at: http://www.gams.com/. Accessed: 16/01/2015.
61
[54]
[55]
[56]
[57]
[58]
[59]
[60]
[61]
[62]
[63]
[64]
[65]
[66]
[67]
[68]
[69]
[70]
[71]
[72]
[73]
62
International Institute for Applied Systems Analysis (IIASA), “BeWhere”.
Available at: http://www.iiasa.ac.at/web/home/research/modelsData/Bewhere/BEWHERE1.en.html. Accessed: 16/01/2015.
R. Bean, J. Wilhelm, “U.S. Alfalfa Exports to China Continue Rapid Growth”,
USDA, 2011.
C. Gopal, M. Mohanraj, P. Chandramohan, P. Chandrasekar, “Renewable energy source water pumping systems—A literature review”, Renewable and
Sustainable Energy Reviews, vol. 25, p. 351–370, 2013 .
Intergovernmental Panel on Climate Change, “Emissions Scenarios”, 2000.
M. Frasconi, “PV solar water pumping technology and system for the conservation of grassland in China: project and analysis of sprinkler and drip irrigation systems”, Mälardalen University, DiVA: diva2:546765, 2012.
Z. Simic, J.G. Havelka, M. Bozicevic Vrhovcak, “Small wind turbines: A
unique segment of the wind power market”, Renewable Energy, vol. 50, p.
1027–1036, (2013).
OpenEI. Available at: http://en.openei.org/wiki/Main_Page. Accessed:
16/01/2015.
Meteonorm. Available at: http://meteonorm.com/. Accessed: 16/01/2015.
Y. Shen, S. Li, Y. Chen, Y Qi, S. Zhang, “Estimation of regional irrigation
water requirement and water supply risk in the arid region of Northwestern
China 1989–2010”, Agricultural Water Management, vol. 128, p. 55–64, 2013.
R.G. Allen, L.S. Pereira, D. Raes, M. Smith, “Crop evapotranspiration - Guidelines for computing crop water requirements”, FAO, 1998.
L. Tanasijevic, M. Todorovic, L.S. Pereira, C. Pizzigalli, P. Lionello, “Impacts
of climate change on olive crop evapotranspiration and irrigation requirements
in the Mediterranean region”, Agricultural Water Management, vol. 144, p.
54–68, 2014.
D.L. Corwin, J.D. Rhoades, J. Simunek, “Leaching requirement for soil salinity control: Steady-state versus transient models”, Agricultural Water Management, vol. 90, p. 165–180, 2007.
T. Markvart, “Solar electricity”, Wiley, 2009.
J.A. Duffie, W.A. Beckman, “Solar engineering of thermal processes”, Wiley,
2006.
M.A. Abella, E. Lorenzo, F. Chenlo, “PV water pumping systems based on
standard frequency converters”, Progress in photovoltaics: research and application, vol. 11, p. 179–191, 2003.
G.P. Kruseman, N.A. de Ridder, “Analysis and evaluation of pumping test
data”, International Institute for Land Reclamation and Improvement, 1994.
T.C. Rasmussen, K.G. Haborak, M.H. Young, “Estimating aquifer hydraulic
properties using sinusoidal pumping at the Savannah River site, South Carolina, USA”, Hydrogeology Journal, vol. 11, p. 466–482, 2003.
P. Steduto, T.C. Hsiao, D. Raes, E. Fereres, “AquaCrop-The FAO crop model
to simulate yield response to water: I. Concepts and underlying principles”,
Agronomy Journal, vol. 101, p. 426–437, 2009.
D. Raes, P. Steduto, T.C. Hsiao, E. Fereres, “AquaCrop-The FAO crop model
to simulate yield response to water: II. Main algorithms and software description”, Agronomy Journal, vol. 101, p. 438–447, 2009.
J.E. Hunink, P. Droogers, “Climate change impact assessment on crop production in Uzbekistan-World Bank study on reducing vulnerability to climate
change in Europe and Central Asia (ECA) agricultural systems”, World Bank,
2011.
[74]
[75]
[76]
[77]
[78]
[79]
[80]
[81]
[82]
[83]
[84]
[85]
[86]
[87]
[88]
[89]
[90]
[91]
[92]
[93]
Solartech. Available at: http://www.solartech.cn. Accessed: 16/01/2015.
Alibaba. Available at: http://www.alibaba.com. Accessed: 16/01/2015.
K. Deb, A. Pratap, S. Agarwal, T. Meyarivan, “A Fast and Elitist Multiobjective Genetic Algorithm: NSGA-II”, IEEE Transactions On Evolutionary Computation, vol. 6, 2002.
Y. Yu, J. Liu, H. Wang, M. Liu, “Assess the potential of solar irrigation systems for sustaining pasture lands in arid regions – A case study in Northwestern
China”, Applied Energy, vol. 88, p. 3176–3182, 2011.
C.J. Vörösmarty, P. Green, J. Salisbury, R.B. Lammers, “Global water resources: vulnerability from climate change and population growth”, Science,
vol. 289, p. 284–288, 2000.
DIVA-GIS. Available at: http://www.diva-gis.org. Accessed: 16/01/2015
WorldClim. Available at: http://www.worldclim.org/. Accessed: 16/01/2015
R.J. Zomer, A. Trabucco, D.A. Bossio, O. van Straaten, L.V. Verchot, “Climate change mitigation: a spatial analysis of global land suitability for clean
development mechanism afforestation and reforestation”, Agriculture, Ecosystems & Environment, vol. 126, p. 67–80, 2008.
H. Haberl, K.H. Erb, F. Krausmann, V. Gaube, A. Bondeau, C. Plutzar, S. Gingrich, W. Lucht, M. Fischer-Kowalski, “Quantifying and mapping the global
human appropriation of net primary production in Earth's terrestrial ecosystem”, Proceedings of the National Academy of Sciences, vol. 104, p. 12942–
12947, 2007.
K. Vadrevu, K. Lasko, “Fire regimes and potential bioenergy loss from agricultural lands in the Indo-Gangetic Plains”, Journal of Environmental Management, vol. 148, p. 10–20, 2015.
M. Ozdogan, “Exploring the potential contribution of irrigation to global agricultural primary productivity”, Global biogeochemical cycles, vol. 25, 2011.
FAO, “Global map of monthly reference evapotranspiration”. Available at:
http://www.fao.org/geonetwork/srv/en/metadata.show?id=7416&currTab=
simple. Accessed: 16/01/2015.
Y. Fan, H. Li, G. Miguez-Macho, “Global patterns of groundwater table
depth”, Science, vol. 339, p. 940–943, 2013.
D. Leclère, P. Havlík, S. Fuss, E. Schmid, A. Mosnier, B. Walsh, H. Valin, M.
Herrero, N. Khabarov, M. Obersteiner, “Climate change induced transformations of agricultural systems: insights from a global model”, Environmental
Research Letters, vol. 9, 2014.
G. Fischer, F.N. Tubiello, H. van Velthuizen, D.A. Wiberg, “Climate change
impacts on irrigation water requirements: Effects of mitigation, 1990–2080”,
Technological Forecasting & Social Change, vol. 74, p. 1083–1107, 2007.
Trading economics. Available at: http://www.tradingeconomics.com/. Accessed: 16/01/2015.
OECD, “OECD Economic Surveys: China 2013”, OECD publishing,
DOI:10.1787/eco_surveys-chn-2013-en, 2013.
IATA-Jet Fuel Price Monitor. Available at: http://www.iata.org/publications/economics/fuel-monitor/Pages/index.aspx. Accessed: 16/01/2015.
Cnagri. Available at: http://en.cnagri.com. Accessed: 16/01/2015.
The Economist, “The drying of the West-Drought is forcing westerners to consider wasting less water”. Available at: http://www.economist.com/news/
united-states/21596955-drought-forcing-westerners-consider-wasting-lesswater-drying-west. Accessed: 16/01/2015.
63