Using solar photovoltaic energy in urban water supply systems: a case study in Kayseri, Turkey

In order to meet the rapidly increasing global energy demands, the search for alternative energy sources is increasing and the greenhouse gases emitted by fossil fuels cause global warming and climate changes (Valipour et al. 2020; Yildiz et al. 2020; Rosen 2021). The use of conventional energy resources in existing water pumping systems used for irrigation, urban/rural water supply, livestock and other purposes is responsible for making a significant contribution to fossil fuel consumption and greenhouse gas emissions (Chowdhury 2018).

Electrical or diesel-powered pumps are widely used in water pumping systems for various purposes such as agriculture and rural/urban water supply. However, the main problems such as non-availability or erratic grid and high diesel pumping costs require a focus on pumping systems using renewable energy sources. The environmental and sustainable development concerns worldwide in recent years have renewed interest in water pumping systems using renewable energy sources (Chandel et al. 2017; Wu et al. 2020). Many renewable energy sources can be used for water pumping. However, the fact that solar energy is available in most places and is unlimited makes it more popular (Verma et al. 2020). Therefore, among the renewable energy sources, solar energy is the most widely used source in water pumping systems (Alshamani & Iqbal 2017). The use of solar energy for water pumping reduces dependence on grid or diesel.

The solar photovoltaic water pumping system (SPVWPS) is based on photovoltaic (PV) technology that converts sunlight into electricity to pump water. A SPVWPS is similar to any other pumping system, with the exception that the power source is solar energy (Foster et al. 2009). Nowadays, one of the most attractive among the applications where PV systems are preferred as an energy source is water pumping systems (Sharma et al. 2020). A solar-powered water pumping system offers a competitive alternative to grid- or diesel-powered water pumping systems. The system consists of a PV array that converts sunlight into electrical energy, a solar pumping inverter (for alternating current (AC)-powered motor), a motor-pump set and a water storage tank (Foster et al. 2009; Ammar et al. 2012; Chandel et al. 2017). The economic benefits and performance of SPVWP systems depend strongly on the global irradiation of the geographic location where the system is installed and many local climatic parameters (Allouhi et al. 2019). The SPVWPS supplies the most water during the hours when the amount of solar irradiation is highest and water is needed the most. The system has advantages such as simplicity, reliability and low maintenance in addition to being economical compared to conventional water pumping systems (Jamil et al. 2012).

The SPVWPS is the best solution for remote areas where grid connection is not possible (Sharma et al. 2020). However, it is a known fact that submersible pumps, used for urban water supply purposes and fully powered by conventional electricity grids, add an additional load to the grids infrastructure (Chowdhury 2018). In developing countries, the energy demand for water pumping is increasing and the electricity grid infrastructure has failed to meet this demand (Mantri et al. 2020). Therefore, reducing the net power consumed from the grid by the pumps is important in terms of sustainability, energy efficiency, economic development and environmental benefits in urban areas such as a city where grid connection is possible. Reduction of conventional energy use and CO₂ emissions in the urban water supply system must be achieved without compromising the water needs of all users.

The SPVWPS used for rural water supply, livestock, irrigation and other purposes has been a subject of immense interest to researchers, policymakers, engineers and investors. Some investigations on the SPVWPS are presented in Table 1. Nayar et al. (1993), Muljadi (1997), Kolhe et al. (2004) and Vongmanee (2005) provided works dealing with reasonably effective PV water pumping systems. Ghoneim (2006) carried out a performance optimization of a PV-powered water pumping system in Kuwait. The system was pumping 12 m³ of water per day using a direct-coupled PV water pumping system in a rural area. The cost of the proposed system was found to be cheaper than that of the conventional fuel system. Padmavathi & Daniel (2011) conducted a case study in the city of Bangalore (India) to reveal the importance of installing solar water pumps in every household in major cities. PV panels ranging from 60 to 500 Wp were found to be sufficient to fill the upper tanks in residential buildings of Bangalore city using solar energy. Jamil et al. (2012) proposed an SPVWPS to meet the water needs of Jamia Millia Islamia University, Faculty of Engineering and Technology in New Delhi, India. The techno-economic analysis of the proposed system was presented and compared with the existing system. Đurin & Margeta (2014) presented a hypothetical example for sustainable energy supply of an urban area located in the southern Mediterranean part of Croatia. The results showed that the proposed innovative solution can reliably and continuously supply the pumps with renewable energy, providing a reliable and continuous water supply to the users. Elías-Maxil et al. (2014) stated that water supply is part of the urban water cycle, reviewed the literature on urban water, and summarized the measures applied in different parts of the world to reduce energy consumption in the urban water cycle. Chandel et al. (2015) presented a comprehensive literature review of using solar energy as a power source in water pumping systems and evaluated the economic feasibility of these systems. They found that SPVWP systems are more economically viable than diesel and conventional electricity grid-based pumping systems for irrigation and water supply in rural, urban and remote areas. Margeta & Đurin (2017) proposed a system consisting of a PV module, an invertor, a pump station, and a reservoir to describe and analyse new and innovative concepts for possible integration of solar energy in urban water supply systems. It was shown that the proposed concepts for local energy production and use are applicable and sustainable. Rodriguez (2018) suggested using the SPVWPS as an alternative to reliance on conventional electricity for groundwater pumping in domestic urban supply in the megacity in the developing world. Mexico City's 528 public urban groundwater wells were selected as a case study and the size and capacity of the SPVWPS were determined. According to a 30-year scenario, the SPVWPS installation was found to be suitable in terms of capacity and the investment was financially feasible. Allouhi et al. (2019) investigated an optimum PV system configuration to meet domestic water needs of five houses located in a remote part of Morocco. Two PV systems with different system configurations: (i) the PV system including a maximum power point tracker direct current (MPPT DC) converter and (ii) the direct coupling PV system were analyzed and evaluated. The results showed that the performance of the PV system with an MPPT DC converter is much better than the direct coupling system.

The literature review shows that many studies have been focused on the technical analysis, increasing the efficiency, economic and environmental benefits of SPVWPS used for irrigation, drinking water supply, livestock and other purposes in rural and remote areas where there is no grid power. However, there are a limited number of studies related to the use of solar PV energy in urban water supply systems. These studies were mostly carried out in detached houses where few people live or in single buildings (Ghoneim 2006; Ramos & Ramos 2009; Jamil et al. 2012; Setiawan et al. 2014; Allouhi et al. 2019). Studies on the use of solar PV energy in urban water supply systems, where a large number of people's water needs are met, are very limited (Padmavathi & Daniel 2011; Rodriguez 2018). It is thought that this study, which investigates the technical and economic benefits of using two different PV system configurations in urban water supply systems, will make an important contribution to filling the gap in this field. This study aims to investigate the technical and economic feasibility of using solar PV energy to reduce the net power consumed from the grid by submersible pumps in urban water supply systems. The Germiraltı catchment area (Kayseri, Turkey), which provides all of its energy needs from the city electricity grid, was chosen as the research area. For this purpose, two different scenarios, ‘hybrid-powered water pumping system’ and ‘grid-connected PV system’, were analyzed and evaluated. The hybrid-powered water pumping system has a separate PV array and an MPPT inverter/controller for each submersible pump. In this system, the submersible pumps are powered by the PV system during the daytime hours when the solar radiation is sufficient, and powered by the city electricity grid when the solar radiation is not sufficient. On the other hand, the grid-connected PV system operates completely independently of the pumping system. The energy generated by the PV system is injected into the grid and the submersible pumps are powered entirely by the electrical grid system. The design, modeling and simulation of both proposed systems were made using the PVsyst v7.02 software. Modeling was made by considering the pumps, pump hydraulic circuits, water needs and heads in the pumping station of the Germiraltı catchment area. The actual data obtained from the SCADA system in the pumping station were compared with the values found as a result of the simulation. The effects of both proposed systems on energy production and the amount of pumped water were determined and evaluated.

SPVWPSs are systems that pump groundwater or surface water for irrigation, urban/rural water supply, livestock and other purposes and provide the power needed to achieve this through PV panels. A typical SPVWPS consists of a solar PV array, a direct current to direct current (DC–DC) converter or a direct current to alternating current (DC–AC) inverter, a motor-pump set, water storage tank, control circuits, suitable mounting structure, manual or automatic tracking system, interconnect cables, protection devices and piping arrangement.

There are three types of SPV water pumping systems: (a) a direct-coupled DC solar photovoltaic water pumping system (SPVDC), (b) an AC solar photovoltaic water pumping system (SPVAC) and (c) SPVWPS with battery storage. In SPVDC, the PV array current and voltage are equal to the armature current and the terminal voltage. The output power of the PV system and pump speed varies with the irradiance level and module temperature. However, if an MPPT is included in the SPVWP system, the output of the PV cells is sampled and the proper resistance (load) is applied and thus the obtained power under all environmental conditions is maximized. In SPVACs, an inverter must be installed between the PV array and the motor. In case the system produces more energy, the unused energy can be stored in batteries depending on the installation or a reservoir can be installed at the pump outlet to store pumped water. Thus, when the solar radiation is non-available, the water stored in the reservoir is used (Trace 2021). By including battery storage in the SPVWP system, solar-generated power is used when consumption is needed, but this increases the cost of the system (Chandel et al. 2015, 2017). The PV modules are connected in series and parallel to supply power to the motor-pump set which is a component of the SPVWPS and produced the required pressures and water flow to water users.

SPVWP systems, which can be used for various purposes, have many advantages, including operational safety, robustness and environmental consciousness (Allouhi et al. 2019; Trace 2021):

• SPVWP systems are durable and are low-cost when compared to fossil fuel-based pumping systems.

• They are reliable and require low maintenance as they don't require an attendant to be present during operation.

• No energy storage is needed as the water can be stored by itself.

• High modularity, PV water pumping systems can adapt to the eventual needs of growth.

• They are environment-friendly and do not cause negative impacts such as water, air or noise pollution.

Kayseri province, located in the Central Anatolia Region of Turkey, is located between 37°45′ and 38°18′ north latitudes and 34°56′ and 36°59′ east longitudes. Kayseri province covers 2.2% of the country's territory with an area of 16,917 km². Its average altitude is 1,054 m. There are 16 districts together with the central districts. Kocasinan and Melikgazi are the central districts. Akkışla, Bünyan, Develi, Felahiye, Hacılar, İncesu, Özvatan, Pınarbaşı, Talas, Sarıoğlan, Sarız, Tomarza, Yahyalı and Yeşilhisar are the other districts (Kayseri Governorate).

Climatic conditions have a significant impact on the performance of PV array and accordingly the solar-powered water pump performance. For this reason, environmental factors such as ambient temperature and irradiation level, which affect the energy converted by the PV generator, should be carefully measured in designing an SPVWP system (Allouhi et al. 2019). Kayseri has cold semi-arid climate characteristics. In the province of Kayseri, the continental climate of Central Anatolia is dominant, with cold and snowy winters and hot and dry summers. Annual average global horizontal irradiation and sunshine hours for Kayseri is 4.36 kWh/m²-day and 7.8 h/day, respectively (Figure 1). Therefore, there is a vast potential for use of solar energy.

The Kayseri Water and Sewerage Administration General Directorate (KASKİ) was established in 1989 to carry out the water and sewerage services of the Kayseri Metropolitan Municipality, to establish all kinds of facilities required for this purpose, to take over the established ones and to operate them from one hand. The KASKI Germiraltı catchment area is one of the 16 large catchment areas used to meet the water needs of Kayseri city center. It is located in the central district of Melikgazi in Kayseri province. The catchment area has 18 wells, 16 of which are active and two are spare. In addition to the wells, the catchment area has a water storage tank, a pumping station and a reservoir. An important part (11.11%) of the water needs of densely populated neighborhoods of Melikgazi and Kocasinan districts is supplied by this catchment area. It is one of the catchment areas in Kayseri that has the highest energy density and energy costs for water supply.

In the KASKI Germiraltı catchment area, 16 active submersible pumps with variable frequency drivers are used to bring groundwater to the surface. These submersible pumps pump water from deep wells to the water storage tank with a volume of 500 m³ located near to the wells and at a height of 3.4 m above ground level. Then, the water in the water storage tank is pumped to the reservoir with a volume of 15,000 m³ by means of centrifugal booster pumps placed at ground level. The water collected in the reservoir is transferred to the water distribution pipelines by gravity flow. Pumping at the pumping station continues for 24 h and all pumps are fully powered from the conventional electricity grid. The submersible pumps used in the deep wells in the Germiraltı catchment area are powered with AC-powered motors, selected according to the water flowrate and head. The motor and pump both are integrated as a single unit in a submersible pump.

PVsyst is a simulation tool developed by the University of Geneva in Switzerland to perform the modeling, simulating, sizing and analysis of PV systems (Ozcan & Ersoz 2019; Uyaver et al. 2020; Yildiz & Yilmaz 2020). PVsyst deals with PV system design such as grid-connected or stand-alone PV systems, solar PV water pumping systems and DC-grid (public transportation). In a PV pumping system, PVsyst determines the flow rate supplied by the pump as a function of the head and available electrical energy generated with the PV array for any running hour in the simulation process. In addition, simulation process manages the situations where the tank is full (limiting the pump's flow at the user's draw and stopping the pump during the rest of the hour), and when the tank is empty (the user's needs cannot be satisfied) (PVsyst 2021). The PVsyst software also shows how different parameters used in the system design affect the results. Software results help to understand the performance and efficiency of the system before installation in the field (Sharma et al. 2020).

It is very important to ensure the optimum system design, performance analysis, technical and economic evaluation of the proposed PV system using simulation software before it is physically installed. There are many softwares for the design and simulation of solar water pumping systems. The PVsyst software is preferred in this study as it is best suited for the SPVWPS design optimization simulation (Sharma et al. 2020) and its international validity and reliability (Yildiz & Yilmaz 2020).

When designing a PV system, it is necessary to use information from the local climate data of the geographic location. In this study, the meteorological data of Kayseri province were obtained from the Meteonorm database embedded in the PVsyst software. Meteorological data of Kayseri province, obtained from the Meteonorm database and used in the simulations, are typical meteorological year (TMY) weather data produced from a total of 21 years of data bank between 1990 and 2011. These data include hourly values of meteorological elements such as solar radiation, ambient temperature, wind speed, etc. It should be noted that the TMY weather datasets used in the simulation do not correspond exactly to the weather data on the site studied. This mismatch between the actual weather data and the TMY weather datasets may, to a certain extent, affect the results obtained in the case of the installation of the proposed PV system.

The tilt angle of the PV modules should be determined to receive maximum solar energy. This tilt angle is the angle at which the loss with respect to optimum is 0%. When the tilt angle is 34°, the loss with respect to optimum is 0%, therefore the simulations were performed for the case where the sun azimuth angle is 0° and the tilt angle is 34°.

The submersible pumps in the KASKI Germiraltı catchment area are fully powered by the conventional electricity grid. This study investigates the feasibility of using solar PV energy to reduce the net power consumed from the grid by the submersible pumps. For this purpose, two different scenarios, ‘hybrid-powered water pumping system’ and ‘grid-connected PV system’, were analyzed and evaluated. In the hybrid-powered water pumping system, the submersible pumps are powered by both the PV system and the electricity grid. While the submersible pumps are powered by PV systems during the daylight hours when the solar radiation is available, they are powered from the electricity grid during the hours when the solar radiation is non-available. On the other hand, in the grid-connected PV system, the PV system is directly connected to the electrical grid and the submersible pumps are fully powered by the grid.

The main purposes of using solar PV energy systems in urban water supply systems are (i) to meet the energy requirement of the submersible pumps entirely with renewable energy sources or to meet some of the energy requirement of submersible pumps with renewable energy sources, thus reducing the power drawn from the electrical network by submersible pumps and (ii) to meet the water needs of the users without interruption. The two proposed scenarios will be evaluated in terms of their ability to achieve these goals and the selection will be made accordingly. Table 2 presents technical specifications of selected PV module for both scenarios and selected inverter for the second scenario.

The submersible pumps are powered by both the PV system and the electricity grid. While the submersible pumps are powered by PV systems during the daylight hours when the solar radiation is available, they are powered from the electricity grid during the hours when the solar radiation is non-available. The schematic of the hybrid system is shown in Figure 2. As can be seen, the hybrid system consists of integrating the PV array and the controller into the existing pumping system.

The PV array and controller have been individually designed for each submersible pump using the PVsyst software. A series of simulations have been performed in which the PV array and inverter size were changed. Thus, the water pumping system fed entirely from the conventional electricity grid has been converted to a hybrid-powered water pumping system.

As the system is converted to a hybrid-powered water pumping system, it accommodates a wide variety of possible operating conditions and includes the complexities of integrated system configurations. Therefore, the system requires an intelligent strategy of energy management. In addition, it is necessary to use an inverter to convert the DC power generated by the PV arrays to AC power since the existing submersible pumps in the facility are AC-powered motors. The inverter/controller with the MPPT algorithm, besides converting DC power to AC power, (i) adjusts AC-powered motor speed depending on the amount of irradiation, (ii) prevents the dry running of the motor, (iii) shuts down the system when the water tank is full (Meah et al. 2008). The simulations of this scenario were performed by including the whole PV and pumping system in the simulation process and taking into account the meteorological data. The nominal powers of the PV arrays and controllers designed for hybrid-powered water pumping system are shown in Table 3. Thus, the total power of the PV array to be installed for the hybrid-powered water pumping system was found to be 670 kWp.

The schematic of the grid-connected PV system is shown in Figure 3. In this system, the submersible pumps are powered by the grid electricity system as they are currently. The energy generated by the proposed grid-connected PV system is injected into the grid. In other words, production and consumption are completely independent of each other. The pumps operate at the specified voltage of the grid and the system does not require an intelligent strategy of energy management. The design process concerns a central PV system sized according to the nominal power that will inject all of the energy consumed by the submersible pumps into the grid. The DC power generated by the PV array system is converted into AC power using an inverter (Padmavathi & Daniel 2011) and is then directed to the transformer. Finally, the AC power is delivered to the grid (Figure 3). The annual energy consumption of the submersible pumps in the Germiraltı catchment area was measured as 2,764,124 kWh (see Tables 5 and 6). Accordingly, the average daily energy consumption of the submersible pumps in the catchment area is around 7,573 kWh. Considering the daily energy consumption of the pumps and system efficiency, a 1,620 kWp PV power plant has to be installed so as to make the submersible pumps fully solar powered. The proposed PV power plant is designed as ground-mounted and will be installed on an area of approximately 13,500 m² as the average land use requirement for PV power plants is 8.3 acres/MWp (Ong et al. 2013). The plant consists of 5,400 units Si-mono PV modules with 300 Wp power and 27 units 50 kW (300–950 V) inverters.

In this section, the parameters used in the performance assessments of the PV system are defined and the cost analysis methodology is explained.

The payback period is defined as the ratio of the initial investment cost of a system to its annual financial gain and is one of the methods used to decide whether a project is feasible (Meah et al. 2008). The payback period is an indicator used to evaluate the economic benefit of a system and simply computes how fast a company will recover its cash investment for a system (Campana et al. 2016; Reniers et al. 2016; Dincer & Abu-Rayash 2019). Payback period is widely used when long-term cash flows are difficult to forecast, because no information is required beyond the break-even point (Coker 2007). Table 4 presents the relevant cost figures in Turkey for the PV systems, which are proposed in the first and second scenarios. The data (cost figures) used in the initial investment cost calculation of the proposed systems were gathered from the ‘2021 Construction and Installation Unit Prices Book’ published by the Republic of Turkey, Ministry of Environment, Urbanization and Climate Change (CSB 2021). All unit prices in the book represent the labor, machinery, material and manufacturing prices based on the country's conditions within the Republic of Turkey. In addition, the electricity tariff for Kayseri is approximately 0.09 $/kWh regardless of time of day.

Cost analysis includes only the installation of a solar PV system since the motor-pump sets, the water storage tank and the reservoir in the existing system will continue to be used in the proposed system. With this analysis, the payback period is calculated and the economic viability of the PV systems is assessed, taking into account the initial investment cost of the installation of the proposed PV system configurations and the energy revenues.

A hybrid-powered water pumping system has been proposed to reduce the net electrical power drawn by existing submersible pumps from the city electricity grid. However, the proposed system should achieve this goal by meeting all the water needs of the users throughout the day. Therefore, the PV system must be able to produce sufficient energy during daylight hours. The total power of the existing submersible pumps in the Germiraltı catchment area is 519 kW (see Table 3). Considering the losses in the PV system, a simulation was made to determine the capacity of the PV assembly that will provide this maximum power. As a result of the simulation, it was decided that a PV system with a power of 670 kWp would be suitable. The characteristics of the suggested hybrid-powered water pumping system are presented in Table 7. The PV arrays designed for the hybrid-powered water pumping system have a total capacity of 670 kWp and include 16 controllers with various nominal powers. The total effective energy at the output of the PV arrays is 1,121.53 MWh. This means that the PV systems can meet approximately 40.6% the total electricity demand of the submersible pumps in a period of a year.

Figure 4 shows the total effective energy at the output of the PV arrays and the total pump operating energy during a typical winter day. The total effective energy at the output of the PV arrays and the total pump operating energy during a typical summer day are presented in Figure 5. The typical winter day refers to the day when the ambient air temperature, the water needs of the users and the amount of solar energy received are typically the lowest. In this study, 19 February was taken as a typical winter day. The typical summer day refers to the day when the ambient air temperature, the water needs of the users and the amount of solar energy received are typically the highest. In this study, 11 July was taken as a typical summer day. Evaluating the results for typical days provides insight into how the proposed system will affect the energy generated and the amount of water pumped throughout the year. The PV array output varies according to solar energy intensity and module temperature. The total effective energy at the outputs of the PV arrays are 1,410 and 2,344 kWh on the typical winter day and the typical summer day, respectively. The submersible pumps consume about 73% (1,031 kWh) of the total effective energy at the output of the PV arrays on a typical winter day and about 85% (1,987 kWh) on a typical summer day. The remaining energy is unused energy as a result of the controller stopping the submersible pumps due to the energy produced by the PV array under the pump producing threshold (EPumpThr) and the low level aspiration of the wells (deep well, drawdown safety) (ELowLev). The unused energy is injected to the grid by the controller, which has an intelligent strategy of energy management. There are slight differences between the results obtained for the typical winter day and the typical summer day. These differences are due to fewer hours of solar radiation being available during winter days and hourly fluctuations in solar radiation during daylight. The PV systems produce energy only during daylight hours whereas the submersible pumps consume energy throughout the day. Therefore, the PV systems produced 19% of the total energy consumption of the submersible pumps on a typical winter day and 29% on a typical summer day. Total effective energy at the output of PV arrays depends on many factors such as solar radiation intensity, sunshine duration, PV cell type, panel placement (latitude, shading), module temperature, ambient air conditions (temperature, humidity, speed, dust, precipitation, etc.), system losses, and DC–AC conversion efficiency of the inverter. The most important of these factors are solar radiation intensity and module temperature. The curve of the total effective energy at the output of the PV arrays show a variation depending on these factors (Figures 4 and 5). The curve reaches a maximum during the hours when the solar radiation is maximum, and the curve shows a decreasing change as the intensity of the solar radiation decreases. On a typical summer day, the curve has a higher maximum point and the area under the curve is larger. This means that the power and total effective energy at the output of the PV arrays are higher on a typical summer day than on a typical winter day. The curve of the total pump operating energy shows an almost similar variation cycle with the curve of the total effective energy at the output of the PV arrays. This can be attributed to the high water needs of the users during the hours when the PV array produces high amounts of energy. These results show that there is a close relationship between the total effective energy at the output of the PV arrays and the total pump operating energy.

While evaluating the proposed systems, it should be taken into account whether the water needs of the users are met as well as the energy generation. Figures 6 and 7 show the water needs of the users, the amount of water pumped (with grid electricity, with PV systems) and the stored water volume in the reservoir during a typical winter day and a typical summer day, respectively. Submersible pumps draw electricity from the city electricity grid during the night. On the other hand, users have low water needs at night. Therefore, at night, more water is pumped into the reservoir for storage. The amount of water stored in the reservoir on a typical winter day increases during the night and reaches its maximum value at 8:00 am. On the other hand, the amount of water stored in the reservoir on a typical summer day increases during the night and reaches its maximum value at 6:00 am. After these hours, when the amount of water in the reservoir reaches its maximum value, the submersible pumps start to be powered from the PV system instead of the grid electricity. Since less energy is generated by the PV systems during the hours when the amount of solar energy received is low, the submersible pumps operate at a lower speed and pump less water. Although the PV system generates the most energy during the hours when the users’ water needs are at their highest, a sufficient amount of water cannot be pumped. Therefore, some of the users’ water needs are met by the water stored in the reservoir. However, the water in the reservoir runs out at 14:00 pm on a typical winter day and at 12:00 on a typical summer day. In addition, after hours when the water in the reservoir is depleted, the amount of water pumped by the PV system is less than the water needs and the users’ water needs cannot be met. Due to the low amount of water pumped during the hours when the submersible pumps are powered by the PV systems, the pumping system cannot meet the users’ water needs not only during some daytime hours but also during some night hours when the water demand is high.

The results obtained in this study showed that the hybrid-powered water pumping system with the suggested capacity (670 kWp) can partially meet the users’ water needs, but not all the needs. The implementation of the following suggestions can enable the hybrid-powered water pumping system to meet users’ water needs to a greater extent or completely:

• The capacity of the PV system can be increased to meet the water needs at a higher amount. However, installing higher-capacity PV arrays will be costlier and require larger installation sites. Therefore, for different capacity increases, the most appropriate capacity should be determined by making technical and economic feasibility. While doing this, it should also be considered whether there are enough sites for the installation of the PV array in the catchment areas.

• If there is enough space for the installation of the PV array, a PV system that is capable of meeting all of the users’ water needs can be installed. This hybrid-powered water pumping system can supply all the necessary energy to the submersible pumps during the daylight hours.

• Even if the PV system can provide all the energy required for the operation of the submersible pumps during the daytime, it cannot provide the required energy during the daytime hours when the solar radiation is not sufficient and during the night hours. During these hours, the required energy will be provided from the city electricity grid. Therefore, an intelligent energy management system must be used to ensure that the hybrid-powered water pumping system operates in optimum conditions.

The total size of the Germiraltı catchment area is 521,933 m². Since 454,702 m² of this area is a green area, it is not suitable for the installation of the PV system. However, an area of 67,231 m² is suitable for the PV system installation and is currently idle. An area of approximately 5,600 m² is required for the installation of the proposed 670 kWp PV system. For these reasons, there is sufficient space in the Germiraltı catchment area for the installation of a PV system that can meet all the users’ water needs.

The payback period of the proposed hybrid-powered water pumping system has been found to be 4.7 years. The payback period is much less than the lifetime of the proposed system.

A PV system completely independent of the pumping system is suggested in this scenario. All of the energy generated by the PV system is injected directly into the city electricity grid. The pumps are powered entirely from the city electrical grid. The amount of pumped water is not affected by fluctuations in solar radiation and other climatic conditions. Thus, the grid-connected PV system reduces the net power drawn by the submersible pumps from the grid or, if it has sufficient capacity, meets all the energy needed by the submersible pumps.

The total power of the existing submersible pumps in the Germiraltı catchment area is 519 kW (see Table 3). Considering the losses in the PV system, a simulation was made to determine the capacity of the PV assembly that will provide all the energy required for the operation of the submersible pumps from the sun. As a result of the simulation, it was determined that a PV system with a power of 1,620 kWp would be suitable. The characteristics of the suggested grid-connected PV system are given in Table 9. A 1,620 kWp PV system with 5,400 PV modules and 27 inverters injects 2,787.8 MWh/year of energy into the grid. The system has an annual average performance ratio (PR) of 85.7% and an annual average daily final yield of 4.71 h/d. The energy consumed by the submersible pumps is 2,764.1 MWh/year. This result shows that the suggested grid-connected PV system can fully meet the total electricity demand of the submersible pumps. The electricity generated in excess of consumption can be sold to the grid.

Figure 8 shows monthly variation of the actual energy consumption of the submersible pumps and the energy injected into the grid by the PV system. The energy injected into the grid by the PV system is a function of insolation. Hence, more energy is generated by the PV system in the summer months when insolation increases. In July and August, when the sunshine duration and the amount of solar energy received are the highest, the amount of energy injected into the grid is the highest. On the other hand, submersible pumps consume the highest energy in August and October. The annual total energy delivered by the grid-connected PV system to the grid meets the total annual energy need of submersible pumps.

Figure 9 shows the monthly variation of performance ratio, capacity factor and final yield for the PV system. Since the capacity factor and the final yield are functions of the energy injected into the grid, both the capacity factor and the final yield increase in summer and decrease in winter. The capacity factor ranged from 13.1% (January) to 25% (August), while the final yield ranged from 3.14 h/d (January) to 5.99 h/d (August). The highest performance ratio was obtained in January (93.7%), and the lowest performance ratio in August (80.4%). The annual average values of the performance ratio, capacity factor and final yield of the proposed system are 85.7%, 19.6% and 4.71 h/d, respectively.

Figure 10 shows the monthly variation of the daily average capture and system losses of the proposed PV system. System loss varied from 0.064 h/d in January to 0.145 h/d in August while capture loss varied from 0.148 h/d in January to 1.316 h/d in August. Both capture and system losses reached lowest values, especially in cold and less sunny months. Annual average daily values of capture and system losses are 0.681 and 0.105 h/d, respectively.

The payback period of the proposed grid-connected PV system has been calculated as 4.1 years. This short payback period shows that installing the grid-connected PV system is economically competitive. Having low operating and maintenance costs makes PV systems even more competitive.

Table 11 provides a comparative summary of the results found for the hybrid-powered water pumping system and the grid-connected PV system. It is important to highlight the following results:

• The hybrid-powered water pumping system can meet only a part of the energy needs of the submersible pumps. However, the grid-connected PV system can meet all of these energy needs.

• The payback period of the grid-connected PV system is shorter than that of the hybrid-powered water pumping system.

• Since the capacity of the grid-connected PV system is higher than that of the hybrid energy water pumping system, the installation site required for the grid-connected PV system is higher. If there are enough suitable sites for the installation of the PV assembly in or next to the catchment areas, a design in which all the energy needs of the submersible pumps will be met from the PV system should be preferred.

• The hybrid energy water pumping system and the grid-connected PV system locally generate the energy needed to operate the submersible pumps. Local energy generation saves energy costs by preventing energy losses due to transport.

• Local energy generation could reduce the stress on the local network. Also is the general network strong enough to absorb centralized solar power generation?

• Adapting or retrofitting small systems with the introduction of renewables for local use often creates additional losses. These losses can be reduced/avoided by designing optimized systems and using an intelligent strategy of energy management.

There are 10,457 power plants in Turkey with a total installed capacity of 99,819.6 MW (TEİAŞ 2021). Due to rapid population growth and industrial development, Turkey's energy needs are increasing. In parallel with the increasing energy need, Turkey's energy production has also increased. Turkey's electricity generation and distribution infrastructure is strong. It has a grid structure that can seamlessly integrate renewable energy sourced electricity generation into fossil sourced electricity generation. On the other hand, as of 2020, Kayseri province has an electric installed power of 2,688 MW. A total of 2,353 GWh energy was distributed to 770,120 subscribers over a total of 26,008 km of grid, 8,136 transformers, 6,785 km of underground cables and 19,223 km of overhead lines (KCETAŞ 2020). The existing grid infrastructure and institutional know-how show that the electricity distribution grid of Kayseri is strong and that the PV system proposed in this study can be easily integrated into the city electricity grid.

There are 8,389 solar power plants (SPPs) in Turkey with a total installed capacity of 7,815.6 MW (TEİAŞ 2021). On the other hand, the province of Kayseri, where this study was conducted, ranks 4th among the provinces in Turkey with an SPP installed power of over 150 MW (Atlas of Energy 2021). In Turkey and in the province of Kayseri, more electricity than the local need generated at SPP power plants during daylight hours is supplied to the city grid. In the night hours when electricity cannot be produced in the SPP power plants, the local electricity demand is met from the existing conventional electricity grid. Existing SPPs in the province of Kayseri work seamlessly in terms of integration into the strong city electricity grid. For this reason, the SPP, which is proposed to be established in the catchment area of Kayseri province in this study, will be able to operate without any problems in terms of integration into the city electricity network, like other SPPs.

This study was conducted to determine the technical and economic feasibility of using solar PV energy to reduce the net power drawn from the grid by submersible pumps in urban water supply systems of Kayseri province. Besides a simulation study, the KASKI Germiraltı catchment area was monitored and real-time data on the amount of water pumped and the energy consumed by the submersible pumps were collected from November 2019 to November 2020. Two different scenarios, ‘hybrid-powered water pumping system’ and ‘grid-connected PV system’, were analyzed and evaluated. The design, modeling and simulation of both proposed systems were carried out using the PVsyst v7.02 software.

The specific energy consumption is calculated as 0.240 kWh/m³. The hybrid-powered water pumping system with a power of 670 kWp, which will provide the total maximum power of the existing submersible pumps in the Germiraltı catchment area, meets approximately 40.6% the total electricity demand of the submersible pumps in a year. The hybrid-powered water pumping system with the suggested capacity (670 kWp) can partially meet the users’ water needs, but not all the needs. A 1,620 kWp grid-connected PV system will be able to meet all of the total annual energy need of submersible pumps. However, the catchment areas of urban water supply systems should have sufficient installation space for the installation of the PV system with the capacity that provides all the energy consumed by the submersible pumps annually. The electricity generated in excess of consumption can be sold to the grid. The grid-connected PV system has the advantages of being simple, of being completely independent from the pumping system, and the fact that the amount of pumped water is not affected by the amount of energy produced by the PV system. The payback period of the hybrid-powered water pumping system is found to be 4.7 years, and the payback period of the grid-connected PV system 4.1 years. From the results obtained in this work, it can be concluded that it is economically competitive to meet the energy requirement of submersible pumps in urban water supply systems with solar PV energy.

The results for Kayseri showed that meeting the energy of submersible pumps with PV systems in urban water supply systems will create significant technical, financial, and environmental gains. On the other hand, Turkey has a high solar energy potential. Therefore, it is of great importance that central and local governments in Turkey expand the use of PV systems in urban water supply facilities. In addition, Turkey ratified the ‘Paris Agreement’ with the Law No. 7335 dated 6 October 2021. This law imposes important environmental and social responsibilities on Turkey. Providing the energy for submersible pumps in urban water supply systems by using solar PV energy instead of fossil fuels will make a significant contribution to fulfilling Turkey's commitments under the ‘Paris Agreement’. In the future, we will extend this work by considering catchment areas of other provinces in Turkey.

All relevant data are included in the paper or its Supplementary Information.