PV water pumping systems for agricultural applications

(1)

Mälardalen University Press Dissertations No. 175

PV WATER PUMPING SYSTEMS FOR

AGRICULTURAL APPLICATIONS

Pietro Elia Campana 2015

School of Business, Society and Engineering Mälardalen University Press Dissertations

No. 175

PV WATER PUMPING SYSTEMS FOR

AGRICULTURAL APPLICATIONS

Pietro Elia Campana 2015

(2)

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.

Fakultetsopponent: Lawrence Kazmerski, University of Colorado Boulder

ISBN 978-91-7485-189-2 ISSN 1651-4238

(3)

Pietro Elia Campana

den 24 april 2015, 09.15 i Delta, Mälardalens högskola, Västerås.

Fakultetsopponent: Lawrence Kazmerski, University of Colorado Boulder

Akademin för ekonomi, samhälle och teknik Mälardalen University Press Dissertations

No. 175

Pietro Elia Campana

den 24 april 2015, 09.15 i Delta, Mälardalens högskola, Västerås. Fakultetsopponent: Lawrence Kazmerski, University of Colorado Boulder

(4)

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

(5)

To my family

M. Peters, Dayton Daily News

(6)

Acknowledgments

This doctoral thesis was conducted at the School of Business, Society and En-gineering, Mälardalen University, Västerås, Sweden. I gratefully acknow-ledge 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 in-valuable 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. stu-dents.

My deep and sincere gratitude goes to my co-supervisor Assoc. Prof. Hai-long 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 the-sis 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, guid-ance, 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 collab-oration with Teroc AB, especially for the hard work and enlightening discus-sions 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, Alexan-der Olsson and Chi Zhang, and at Institute of Water Resources and

(7)

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 Magnus-son, 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 Inter-national 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.

(8)

Summary

Grassland and farmland degradation is considered as one of the worst envi-ronmental and economic threats for China. The degradation process nega-tively affects food and water security, economy, society and climate changes.

Photovoltaic water pumping (PVWP) technology for irrigation is an inno-vative 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 re-quirement (IWR) and crop response to water have been presented and inte-grated. 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 be-tween 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 sys-tems in China are large. Nevertheless, the feasible and optimal locations are

(9)

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 tech-nology can be applied for irrigation purposes all over the world both for off- and on-grid applications providing several economic and environmental ben-efits.

(10)

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 konst-bevattning är en innovativ och hållbar lösning för att hindra nedbrytningen av gräsområden och samtidigt gynnar den bevarandet av jordbruksområden, sär-skilt inom avlägsna områden i Kina. PVWP-teknologi i kombination med vat-tenbesparande tekniker för konstbevattning och ett hållbart bruk av grundvat-tenresurserna kan medföra flera fördelar – såsom ökad gräsproduktion, en minskning av vind- och nederbördserosionen och ökade inkomster och för-bättrade levnadsvillkor för jordbrukarna.

Syftet med denna doktorsavhandling är att överbrygga rådande kunskaps-luckor, 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 nyttjan-det 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 av-seende på utformning och en ny procedur för optimering, baserad på en gene-tisk 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 kostnads- effektiva lösningarna för konstbevattning och för att utvärdera projektets lön-samhet. De möjliga fördelarna som genereras genom implementeringen av PVWP-systemet har belysts såväl som effekterna av de känsligaste para- metrarna, 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 vat-tenpumpar och vindkraftsdrivna vatvat-tenpumpar, på grund av ett höga och säkra

(11)

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 foderprodukt-ionen. 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 behö-ver 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 mil-jömässiga fördelar.

(12)

List of papers

This doctoral thesis is based on the following papers, which are referred to in the text by their Roman numerals:

I. P.E. Campana, H. Li, J. Yan, “Dynamic modelling of a PV pumping sys-tem with special consideration on water demand”, Applied Energy, vol. 112, p. 635–645, 2013.

II. P.E. Campana, A. Olsson, H. Li, J. Yan, “An economic analysis of pho-tovoltaic water pumping irrigation systems”, International Journal of

Green Energy, 2015 (Article in press).

III. P.E. Campana, H. Li, J. Zhang, R. Zhang, J. Liu, J. Yan, “Economic opti-mization of photovoltaic water pumping systems for irrigation”, Energy

Conversion and Management, vol. 95, p. 32–41, 2015.

IV. P.E. Campana, H. Li, J. Yan, “Techno-economic feasibility of the irriga-tion system for the grassland and farmland conservairriga-tion in China: photo-voltaic vs. wind power water pumping”, manuscript for journal consider-ation.

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}

Confer-ence 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 Manage-ment. Paper IV is currently a manuscript for journal consideration.

Licentiate thesis:

P.E. Campana, “PV water pumping systems for grassland and farmland con-servation”, Mälardalen University Press, Licentiate thesis 172, ISBN 978-91-7485-127-4, 2013.

(13)

The author has also contributed to the following relevant publications which do not constitute part of this doctoral thesis:

I. A. Olsson, P.E. Campana, M. Lind, J. Yan, “PV water pumping for carbon sequestration in dry land agriculture”, Energy Conversion and

Manage-ment, DOI:10.1016/j.enconman.2014.12.056, 2015 (Article in press).

II. A. Olsson, P.E. Campana, M. Lind, J. Yan, “Potential for carbon seques-tration and mitigation of climate change by irrigation of grasslands”,

Ap-plied Energy, vol. 136, p. 1145–1154, 2014.

III. J. Zhang, P.E. Campana, J. Liu, R. Zhang, J. Yan, “Model of evapotran-spiration and groundwater level based on photovoltaic water pumping system”, Applied Energy, vol. 136, p. 1132–1137, 2014.

IV. 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, V. 2014.U. Desideri, P.E. Campana, “Analysis and comparison between a concen-trating solar and a photovoltaic power plant”, Applied Energy, vol. 113, p. 422–433, 2014.

VI. 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.

VII. P.E. Campana, A. Olsson, C. Zhang, S. Berretta, H. Li, J. Yan, “On-grid photovoltaic water pumping systems for agricultural purposes: compari-son of the potential benefits under three different incentive schemes”,

Pro-ceedings of the 13th_{World Renewable Energy Congress (WREC XIII),}

August 3–8 2014, London, United Kingdom.

VIII. 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”,

Pro-ceedings of the 6th_{International Conference on Applied Energy (ICAE}

2014), May 30–June 2 2014, Taipei, Taiwan.

IX. C.L. Azimoh, B. Karlsson, F. Wallin, P. Klintenberg, S.P. Chowdhury, S. Chowdhury, P.E. Campana, “The energy loss in guiding against equip-ment theft in Thlatlaganya Village, South Africa”, Proceedings of the 5th

International Conference on Applied Energy (ICAE 2013), July 1–4 2013

(14)

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

(17)

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

(18)

List of tables

Table 1: AquaCrop parameters for Alfalfa yield simulation. ... 22 Table 2: Genetic algorithm (GA) parameters. ... 24 Table 3: Decision variables and corresponding design spaces. ... 25 Table 4: Spatial data used for the assessment of the technically suitable

grassland areas for PVWP systems for irrigation. ... 27 Table 5: PVWP system components and characteristics. ... 28 Table 6: Measurements carried out during the tests and the corresponding

instruments, resolutions, and accuracies... 29 Table 7: Existing and optimized PVWP systems characteristic

(19)

List of principal symbols and acronyms

AC Alternate current

ADB Asian Development Bank APV PV array area (m2)

c Scale factor (m/s) CF Cash flow ($)

CWR Crop water requirement (mm or m3_{/ha per unit of time)}

D Hydraulic diffusivity (m2_/hour)

DC Direct current DM Dry matter

DWP Diesel water pumping ea Actual vapour pressure (kPa)

es Saturation vapour pressure (kPa)

Es Monthly average daily solar irradiation (kWh/m2/day)

ETc Evapotranspiration in cultural conditions (mm or m3/ha per unit of

time)

ETc,adj Actual evapotranspiration (mm or m3/ha per unit of time)

ETo Reference evapotranspiration (mm or m3/ha per unit of time)

Ew Specific monthly average daily energy yield of the wind turbine

(kWh/kWr/day)

f System failure

f(v) Probability density function of the wind speed (%) FAO Food and Agriculture Organization

fm Matching factor

fuelann Annual fuel cost ($)

g Acceleration due to gravity (9.8 m/s2₎

G Soil heat flux density (MJ/m2_{per unit of time)}

GA Genetic algorithm

Gb,t Beam solar radiation on the tilted surface (kW/m2)

Gd,h Diffuse horizontal solar radiation (kW/m2)

Gd,t Diffuse solar radiation on the tilted surface (kW/m2)

Gg,h Global horizontal solar radiation (kW/m2)

Gg,t Global solar radiation on the tilted surface (kW/m2)

Gr,t Reflected solar radiation on the tilted surface (kW/m2)

Hirr Required head to operate the irrigation system (m)

Ho Operational hydraulic head (m)

(20)

Hr Reference hydraulic head (m)

i Imaginary number ICC Initial capital cost ($)

ICCann Annualized initial capital cost ($)

IIASA International Institute for Applied Systems Analysis IPCC Intergovernmental Panel on Climate Change ir Interest rate (%)

IRR Internal rate of return (%)

IWR Irrigation water requirement (mm or m3_{/ha per unit of time)}

IWRHR Institute of Water Resources and Hydropower Research

IWRj,m Monthly average daily irrigation water requirement for the j-th

crop in the m-th month (m3_/ha/day)

IWRPA Institute of Water Resources for Pastoral Areas

IWRt,m Total monthly average daily irrigation water requirement in the

m-th monm-th (m3_/ha/day)

j j-th irrigated crop k Weibull shape factor

K0 Zero-order modified Bessel function

Kc Cultural coefficient

Ks Water stress coefficient

KTH Kungliga Tekniska Högskolan (Royal Institute of Technology) Ky Yield response factor

LCC Life cycle cost ($)

LR Leaching requirements (%) m m-th month

MPPT Maximum power point tracker N Project lifetime (year)

n Year of the investment (year)

NOCT Nominal operating cell temperature (°C) NPP Net primary productivity

NPV Net present value ($)

omrann Annual operation, maintenance and replacement cost ($)

p Velocity-power proportionality number Pann Annual profit ($)

PBP Payback period (year) pc Pumping cycle (hour)

Pe Effective precipitation (mm or m3/ha per unit of time)

pf Specific forage price ($/tonne DM)

Pp,o Operational pump power (kW)

Pp,PVWP Photovoltaic water pumping power peak (kWp)

PPV PV array power output (kW)

Pr Wind turbine rated power (kWr)

Pr,WPWP Wind power water pumping rated power (kWr)

(21)

Pv Power produced by the wind turbine at wind speed v (kW)

PV Photovoltaic

PVWP Photovoltaic water pumping PW Present worth ($)

Qo Operational water flow (m3/hour)

Qr Reference water flow (m3/hour)

r Distance from the pumping well (m) Rann Annual revenue ($)

Rn Net radiation at the soil surface (MJ/m2per unit of time)

S Storativity s Drawdown (m) sh Hourly drawdown (m)

SH Static head (m)

smax Maximum drawdown (m)

STC Standard test conditions

T Aquifer transmissivity (m2_/hour)

t Time (hour)

Ta Ambient temperature (°C)

Tcell Photovoltaic cell temperature (°C)

TDH Total dynamic head (m)

TSTC Temperature at standard test conditions (25 °C)

v Average wind speed (m/s)

v2 Average wind speed at 2 m above the ground (m/s)

vi Cut-in wind speed (m/s)

Vmp Voltage at maximum power point (V)

vo Cut-out wind speed (m/s)

Vp,d Daily pumped water volume (m3/day)

vr Rated wind speed (m/s)

Vs,d Daily sustainable pumped water volume (m3/day)

WPWP Wind power water pumping WT Wind turbine

x Total number of irrigated crops Ya Actual yield (tonne DM/ha/year)

Ym Maximum yield (tonne DM/ha/year)

Greek symbols α Solar altitude (˚) αC β γ γs γsu δ Δ

Temperature coefficient of the PV power (%/°C) Tilt angle (°)

Psychrometric constant (kPa/°C) Solar azimuth angle (°)

Surface azimuth angle (°) Declination angle (°)

(22)

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 (°)

(23)

1

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 mil-lion 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 degrada-tion, 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 deg-radation and desertification in China are not only restricted to grassland but also affect farmland, causing crop yield reduction and aggravating the coun-try’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 coun-tries [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, pho-tovoltaic (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 ap-plication 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).

(24)

2

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 condi-tions 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 ir-rigation), can lead to a substantial reduction in energy and water requirements of the pastoral and agricultural sectors. PVWP is therefore a promising tech-nology 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 sys-tems, 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 sys-tems and performance improvement of the solar array), and technical and eco-nomic 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 re-source availability, and crop yield under different water supply conditions. System failures and corresponding economic losses still occur due to the in-adequacy 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 ment (e.g., IWR and water resources) and how the system affects the environ-ment (e.g., pumped water and crop yield). Thus, the first two research ques-tions are: how does the variation in IWR during the irrigation season af-fect PVWP systems design and efaf-fectiveness? (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 eco-nomic, technical, and environmental constraints that can negatively affect the

(25)

3 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 pump-ing 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, con-ducted 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 envi-ronmental suitability of PVWP technology for pastoral and agricultural pur-poses, specifically for restoration of degraded grassland and farmland conser-vation.

Corresponding to the above-listed research questions, specific objectives include the following:

 providing a better understanding of the matching between PVWP sys-tem 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, ana-lysing the negative environmental effects, and direct and indirect ben-efits (O3);

 providing a reliable methodology to assess technically suitable and optimal locations for implementing PVWP systems for irrigation (O4).

(26)

4

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. Neverthe-less, 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:

 developing an integrated framework to design and optimize PVWP sys-tems for irrigation. Special consideration is given to the matching between

(27)

5 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 comprehen-sive 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 de-velopment 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.

(28)

6

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 sub-humid areas due to climatic and anthropogenic factors [15]. A significant pro-portion (27.3%) of the Chinese national land area is affected by desertifica-tion, 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 ir-rigation, over cultivation, and pollutant discharges) [16, 17], whereas the sec-ond 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 al-ready fragile environmental and ecological situation.

While preventing desertification includes integrated water and land man-agement 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 stop-ping erosion, such as by enriching soil with nutrients and through afforestation and reforestation practices.

(29)

7

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 fer-tility, vegetation biomass, biodiversity, and water storage [22]. The effective-ness 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 rec-ommended 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 farm-ing-pastoral areas.

Recently, the application of PVWP systems for grassland irrigation in China has attracted more and more attention from the scientific community as

(30)

8

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 com-ponents, but allows harvesting of 30–40% more solar irradiation than the fixed system. The controller depends on the pump motor type, whether direct cur-rent (DC) or alternate curcur-rent (AC). In the first case, the AC pump is connected to the PV array through a DC/AC inverter that transforms the DC power pro-duced 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 sys-tem 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.

(31)

9 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 fea-sibility of PVWP systems for irrigation, concluding that there are no techno-logical 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 irri-gation. 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 independ-ent from fossil fuels and so overcomes all related disadvantages: fuel availa-bility, 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 ab-sence 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 electri-cal 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 irri-gation, 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 pur-pose 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.

(32)

10

Recent research on PVWP systems has mainly focused on system model-ling. 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 un-der 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 imple-mentation of PVWP systems for grassland protection, with installation of a 2 kWp pilot system in Qinghai province [23]. The project demonstrated the

tech-nical and economic feasibility of PVWP systems for preventing grassland deg-radation, farmland conservation, and poverty alleviation in rural areas. The following aspects related to the implementation of PVWP systems in grass-land were addressed: overexploitation of water resources, institutional and fi-nancial barriers, comparison with traditional off-grid water pumping systems, and guidelines for identifying the most suitable locations. Suitability was eval-uated through spatial analysis of land cover, terrain slope, rainfall, tempera-ture, 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 fo-cused 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 considera-tion 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 biodiver-sity in Xilamuren grassland, Inner Mongolia. Results indicated that, even

(33)

11 though irrigation can improve both grass production and biodiversity, the halt-ing 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 tech-nology 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 dy-namic operation of PVWP systems. In particular, there are no previous de-tailed 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 sys-tems, 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 as-sessed the profitability of PVWP systems for grassland irrigation. Previous studies considered only forage sales for a specific location with a specific for-age yield as PVWP system revenue, thus failing to provide a comprehensive analysis of profitability. There is thus the need to analyse PVWP system suit-ability, including all potential benefits. Previous assessments of suitable and optimal locations for PVWP systems in China also disregarded spatial param-eters, such as groundwater availability, groundwater depth, and potential for-age yield; they also did not take into account costs and benefits across the entire forage supply chain.

(34)

12

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 irra-diation, IWR, and the effect of climate change on IWR. This thesis also pre-sents 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 de-signing and simulating PV systems. Unlike most other programs, it also in-cludes 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 tech-nical 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 Ag-riculture Organization (FAO) of the United Nations. Dynamic modelling of

(35)

13 the integrated system is used to validate the design process through the match-ing 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 In-ner 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 algo-rithm (GA) for PVWP irrigation systems. The optimization of the PVWP sys-tem 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 ex-tremely easy.

The economic aspects of the PVWP systems were also thoroughly investi-gated. ICC and LCC were used to compare different water pumping technol-ogies, and net present value (NPV), internal rate of return (IRR), and payback period (PBP) were used to evaluate the capacity to produce benefits. The en-vironmental 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 tech-nology 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

(36)

14

working efficiency of the inverter and pump. The water demand model esti-mates the IWR and, together with the hydraulic head model, is crucial to de-sign 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 assess-ment 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.

(37)

15 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=_𝑓𝑓 0.0027 𝑇𝑇𝑇𝑇𝑇𝑇 𝑚𝑚[1 − 𝛼𝛼𝑐𝑐(𝑇𝑇𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐− 𝑇𝑇𝑆𝑆𝑆𝑆𝑆𝑆)]𝜂𝜂𝑝𝑝max𝑚𝑚 𝐼𝐼𝐼𝐼𝐼𝐼_{𝑡𝑡𝑝𝑚𝑚}(𝐼𝐼𝑝𝑝𝑆𝑆𝑆𝑆) 𝐸𝐸𝑠𝑠𝑚𝑚 (1) 𝑃𝑃𝑟𝑟𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 =0.0027 𝑇𝑇𝑇𝑇𝑇𝑇_𝑓𝑓 𝑚𝑚𝜂𝜂𝑝𝑝 𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚 𝐼𝐼𝐼𝐼𝐼𝐼_{𝑡𝑡𝑝𝑚𝑚}(𝐼𝐼𝑝𝑝𝑆𝑆𝑆𝑆) 𝐸𝐸𝑤𝑤𝑚𝑚 (2) 𝐼𝐼𝐼𝐼𝐼𝐼_{𝑡𝑡𝑝𝑚𝑚}(𝐼𝐼𝑝𝑝𝑆𝑆𝑆𝑆)= ∑ 𝐼𝐼𝐼𝐼𝐼𝐼_{𝑗𝑗𝑝𝑚𝑚}(𝐼𝐼𝑝𝑝𝑆𝑆𝑆𝑆) 𝑥𝑥 𝑗𝑗𝑗𝑗 𝑚𝑚 = {𝐽𝐽𝑚𝑚𝐽𝐽𝑝 𝐽𝐽𝐽𝐽𝐽𝐽𝑝 𝐽 𝑇𝑇𝐽𝐽𝐷𝐷} (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/s}2_{), and the conversion between}

Joule and kWh (1/(3.6*106_{)) to calculate daily hydraulic energy; f}_m_{is the}

matching factor assumed to be equal to 0.9 (the matching factor can be ad-justed 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

temper-ature (°C); TSTCis 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); Esis the monthly average daily solar irradiation hitting

the PV array (kWh/m2_{/day); E}_w_{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

sustaina-ble pumped water volume Vs,d to avoid overexploitation of groundwater

re-sources. 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:

(38)

16

𝑇𝑇𝑇𝑇𝑇𝑇 = 𝑆𝑆𝑇𝑇 + 𝑠𝑠𝑚𝑚𝑚𝑚𝑚𝑚+ 𝑇𝑇𝑖𝑖𝑖𝑖𝑖𝑖+ ∑ 𝜆𝜆 + ∑ 𝜆𝜆𝜆 𝜆(4)

where SH is the static head (m); Hirr is the required head to operate the

irriga-tion system (m); λ and ξ are continuous and concentrated head losses (m), respectively; Hirrhas 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

cal-culated 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]:

𝐸𝐸𝑤𝑤= 𝑡𝑡 𝑡 𝑡𝑡𝑣𝑣 𝑣𝑣𝑟𝑟 𝑣𝑣𝑖𝑖 𝑓𝑓(𝑣𝑣)𝑑𝑑𝑣𝑣 + 𝑡𝑡𝑡𝑡𝑖𝑖𝑡 𝑓𝑓(𝑣𝑣)𝑑𝑑𝑣𝑣 𝑣𝑣𝑜𝑜 𝑣𝑣𝑟𝑟 𝜆 𝜆(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: 𝑡𝑡𝑣𝑣= 𝑡𝑡𝑖𝑖𝑣𝑣 𝑝𝑝_{− 𝑣𝑣} 𝑖𝑖𝑝𝑝 𝑣𝑣_𝑖𝑖𝑝𝑝− 𝑣𝑣_𝑖𝑖𝑝𝑝 𝜆(6) 𝑓𝑓(𝑣𝑣) =𝑘𝑘_{𝑐𝑐 (}𝑣𝑣_𝑐𝑐)𝑘𝑘𝑘𝑘𝑒𝑒𝑘(𝑣𝑣𝑐𝑐)𝑘𝑘 𝜆(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 horizon-tal solar irradiation varies greatly from 3–6 kWh/m2_{/day, depending on}

loca-tion 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].

(39)

17 To assess available solar irradiation and wind speed at specific sites, Mete-onorm [61] and Open Energy Information [60] were used as a climatic data-base and spatial dataset, respectively. PVsyst [46] and our developed codes were used to calculate solar irradiation hitting the PV array depending on ori-entation angle and array technology. The optimization of PV array oriori-entation 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 hu-midity), crop characteristics (type, variety, and development stage), manage-ment, and environmental conditions. Typically, the computation of IWR can be performed by sequentially calculating reference evapotranspiration, evap-otranspiration under standard cultural conditions, and effective rainfall [62].

Reference evapotranspiration ETo(mm/day) was calculated with the FAO

Penman–Monteith equation [63]: 𝐸𝐸𝐸𝐸𝑜𝑜 =

0.408 𝛥𝛥 (𝑅𝑅𝑛𝑛− 𝐺𝐺) + 𝛾𝛾 900_𝐸𝐸_𝑎𝑎_{+ 273 𝑣𝑣}2 (𝑒𝑒𝑠𝑠− 𝑒𝑒𝑎𝑎)

𝛥𝛥 + 𝛾𝛾 (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/m}2_{/day); T}_a_{is the monthly}

average daily air temperature (°C); Δ is the saturation slope of the vapour pres-sure 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):

𝐶𝐶𝐶𝐶𝑅𝑅 = 𝐸𝐸𝐸𝐸𝑐𝑐− 𝑃𝑃𝑒𝑒= 𝐸𝐸𝐸𝐸𝑜𝑜𝐾𝐾𝑐𝑐− 𝑃𝑃𝑒𝑒 (9)

where Kc is the cultural coefficient dependent on crop growth stage. The

ef-fective 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: 𝐼𝐼𝐶𝐶𝑅𝑅 =_ƞ 𝐶𝐶𝐶𝐶𝑅𝑅