The Constraint of the Eco-Friendly Rooftop PV System Implementation Based on the Techno-economic and Regulatory Impact Analysis: Case Study of the Ontowiryo Mosque in Purworejo, Central Java, Indonesia

PV system is an eco-friendly option to meet the need for energy due to its lower carbon footprint when compared to the fossil-fueled power plants. In this research, we performed techno-economic and regulatory impact analysis on a rooftop PV system on a mosque in Purworejo, Indonesia under a net metering mechanism. The use of PV system on the mosque rooftop can reduce its entire carbon emission from its annual energy usage, which equals to 4 tonnes of carbon dioxide equivalent. However, the economic feasibility of the PV systems, measured by the NPV of the electricity bill saving, shows negative values for all PV system conﬁguration, deﬁned by the type and number of panels. This low ﬁnancial attractiveness is resulted from several unsupportive regulations: the subsidized electricity tariﬀ, the minimum grid electricity usage limit and the less rewarding net metering scheme. The abolishment of minimum usage limit incurs the greatest increase of NPV to the PV system that is designed to entirely supply the load demand. While the implementation of higher electricity tariﬀ incurs the highest increase of NPV for the PV systems with partial supply scenario. This study shows that at the current electricity tariﬀ, costs of components, and the implemented regulations around the adoption of PV system, it is ﬁnancially unfeasible to install a PV system on the mosque rooftop.


Introduction
In order to tackle the climate change, Indonesia has planned to increase its energy mix for electricity production from renewable energy by 23% in 2025 [1]. Until the first semester of 2020, DEN [2] reported that the share of renewable energy for electricity generation has only reached 10.9%. Renewable energy, including solar energy, is eco-friendly and can be utilized for generating power with carbon footprint lower than those of the fossil fueled power plants. Tawalbeh et al. [3] estimated that the carbon footprint from PV systems was in the range of 14-73 g CO2-eq/kWh, which is only 2 to 10% of the carbon emitted to generate 1 kWh of electricity by burning the oil.
On the other hand, through their studies, IESR [4] suggested that Indonesia can achieve the energy transition goal by utilizing the distributed generation of solar power in the form of rooftop and off grid PV system. It is estimated that there is at least a potential of 2 GWp rooftop solar PV in the residential market of Jakarta and Surabaya. Hence, it is implied that a greater potential is expected when more cities and other type of buildings are considered.
For this reason, this paper aims to explore the design and feasibility of rooftop PV system. The case study considered in this paper is the rooftop solar PV of mosque, specifically the Ontowiryo Mosque in Purworejo, Central Java [5]. The choice of mosque as the case study is also supported by the fact that Indonesia is one of the countries with the largest population of Muslims. To support their activities, Ministry of Religion [6] reported that Indonesia has more than half a million of mosques and prayer halls with various sizes. Ghazali et al. [7] stated in their book that the use of clean energy in such facility is also relevant ethically and morally since the preservation of the nature and environment is one of the teachings of Islam.
The commandment for humanity to utilize solar energy is also implicitly stated in the Holy Quran, the holy book of Islam. This order is stated in the first verse of Chapter 108 that says: "Surely We have given thee abundance of good" [8]. Through this verse, the muslims believe that God Almighty has given an abundance of goodness to humanity through the manifestation of the universe that includes the earth, sun and stars, and so on to provide for the sustainability of the humanity. Furthermore, some of the main purposes of the sun is to illuminate the earth with its rays and provide an immense source of energy. For that reason, until now, human- ity keeps exploring the benefits that can be utilized from the use of solar energy.
This argument also serves as one of the underlying reasoning for conducting this research, that is, to explore the utilization of solar energy through the PV system on the rooftop of Ontowiryo Mosque in Purworejo, Indonesia for the lighting purposes of the mosque and the neighborhood.
There has been a significant number of researches on the implementation of rooftop solar PV for mosques, as will be briefly discussed in this paper. Rashid et al. [9] analyzed mosque rooftop PV system could reduce the electricity bill annually by 47% in Malaysia. The result shows that the payback period of the PV system's initial investment is reached in 13 years. Similar study was performed in Kuwait which resulted the same payback period. Almutairi [10] investigated the financial feasibility of rooftop solar PV system at 1400 mosques in Kuwait is studied in 2018.
The PV systems are designed to supply the connected load and shave the peak load during the day. Both studies however did not consider the present value of future cashflows. Thus, the benefit of the rooftop PV systems is subject to overestimation and the real payback period may be achieved at a later date.
In 2019, Elshurafa et al. [11] conducted a more comprehensive pilot study of mosque rooftop PV system in Saudi Arabia. The PV system studied is connected to the grid for different scenarios: no support policy and with net-metering mechanism which enables the compensation for the exported surplus-generation to the grid. The study finds that the analyzed system is financially attractive for both scenarios. The implementation of net-metering reduces the cost incurred for fulfilling electricity needs 22% lower than the rooftop PV system with no support policy scenario.
As of 2018, the government of Indonesia [12] have ratified the latest regulation on the net metering scheme for solar rooftop. Therefore, this paper aims to perform a similar study to the previously mentioned related papers, that is to explore the design and feasibility of mosque rooftop PV system with a case study location in Indonesia.
To that end, we present for the first time a detailed techno-economic analysis of mosque policymakers on the effectiveness of the current supporting policy for the implementation of the rooftop PV system for a specific type of end-user.

Methods
To achieve the research goals, three main activities are conducted in this research: data acquisition, rooftop PV system design and simulation, and design result analysis. The analyses in each activity are done with using codes written in MATLAB.

Data Acquisition
The data acquisition process prepares the required data for the PV system design process, which includes site condition data and hourly electricity load profile data. The site condition data are obtained from Solcast [13] Toolkit which provide a comprehensive set of hourly site condition data that includes air temperature, cloud opacity, solar irradiance (DHI, DNI and GHI), sun azimuth angle, sun altitude angle, wind speed, relative humidity and precipitable water.
The hourly load profile data is ideally generated from historical data, by taking a sample of electricity consumption in one day. This method is unavailable for the considered case study as the mosque is currently under construction. Therefore, the load profile of the mosque is estimated by calculating the lighting and electronic equipment that is possibly used for certain activities and at certain times. To conclude, this research work with 8784 hourly data to simulate a full year operation of rooftop PV system.

Rooftop PV System Design
The solar irradiance received by the rooftop area arrives in three forms: direct and diffused irradiance and the irradiance reflected by the ground [14]. The solar irradiance data obtained in the previous activity are used to calculate each value of these forms, which are then summed to determine the irradiance received by the rooftop and, hence, the estimated yield of solar energy on the mosque roof.
This paper studied a grid connected PV system. The optimal number of solar panels and inverter capacity in the PV systems are determined by modeling the PV system using the solar radiation data and the electrical load profile as the input. This paper employs two design scenarios: complete and partial fulfillment of load demand by the PV systems which will be referred as the first and second scenario respectively in this paper.
In Indonesia, a monthly minimum limit of electricity usage as much as 40 hours of consumption at maximum power limit are imposed to the grid-connected user [15].
Thus, the partial scenario entails a combination of the PV system and the grid to supply the load demand. The grid supplies the load demand under the minimum limit, while the PV system supply the excess.
At each hourly time steps, if the electricity supply from the PV system is smaller than the load demand, the electricity load will be met by electricity from the PV system while importing the mismatch from the grid. Thus, the load demand met by the PV system during this condition is equal to the PV system generation. Vice versa, when the electricity supply from the PV system is greater than the electricity load at a certain time, the electricity load will be completely supplied by the PV system and excess electricity from the PV system will be exported to the grid.

Design Result Analysis
The design result analysis incorporates the calculation of carbon emission reduction and financial feasibility study of the PV system. The benefit of carbon emission reduction is then translated in monetary terms by using the social cost of carbon emission so that it can be easily compared with the financial feasibility study. The financial feasibility study was conducted using the analysis of Net Present Value (NPV) and Levelized Cost of Electricity (LCOE).
The NPV is calculated by using the Equation 1 with R t represents the cashflow, in and out, in year t with a discount factor i. The cash outflow accounts the initial investment and annual operation and maintenance cost. While the cash inflow is defined as the annual electricity bill reduction. Thus, the NPV value in this paper regards the present value of the electricity bill savings [14]. This metrics also indicate the recovery rate of capital expenditure (CAPEX), with zero NPV means full capital expenditure recovery. LCOE is calculated by using Equation (2) with OPEX is the operational and maintenance expenditure, and P V util is the energy generated by the PV system that can be utilized by the mosque. The financial feasibility analysis is followed by a sensitivity analysis on the impact of different regulations implementation. The change of the PV systems' NPV is used as the metric to measure the impact of the change in regulation.

Site Information
The Ontowiryo Mosque is located on the south side of Purworejo district, Central Java, at 109.96°Longitude, -7.85°Latitude. The specification of the roof is shown in figure 1. The mosque is connected to the grid with a maximum power limit of

Financial and Operational Assumption
To ensure the validity of the data and to simplify the comparison of different types of solar panels, the data for solar panels are taken from a website of renewable en- The 2019 PLN statistics [19], which provide data on the average electricity tariff

PV System Configuration
The site condition analysis shows that the solar irradiance potential in the case study location is at 1971 kWh/year.m2 or equals to 5.4 daily Equivalent Sun Hours (ESH). This potential is achieved with an optimal configuration of solar panel tilt angle and azimuth at 10°and 6°, respectively, as shown in figure 4. Consequently, as the existing roof orientation is different from the optimal configuration, the potential of solar irradiance in each sides of the roof, west and east, are at 4.9 and 3.5 ESH.
When entire area of the roof is considered, the total potential for the west and east side of the roof are respectively at 263.24 and 141.47 MWh annually.

Performance Analysis
The result of the performance analysis shows that the PV system efficiency ranges from 15.6% to 19.37%, and energy yield ranges from 1216 to 1263 kWh/kWp in a year. PV system using REC AA series panel achieved the highest system efficiency and solar energy yield due to their high panel efficiency.
The result also shows that the capacity of the PV system ranges from 4020 Wp to 4615 Wp to completely meet the load demand. With this capacity, at least around 25% of the generated electricity is lost in the grid during the export and import process. The net-metering scheme in Indonesia only compensates 65% of the exported energy. Hence, a smaller amount generated electricity leads to a smaller amount of exported electricity in the grid, and, thus smaller losses.
For the same reason, the PV system in the second design scenario has a lower grid Figure 3 The average electricity tariff of social sector users in Indonesia. Adapted from [19]. loss at around 16%. The annual minimum usage limit of the mosque is at 2640 kWh, which is also equal to half of the mosque's annual load demand. Thus, these PV systems only employ half as much number of panels with capacities ranging from 1950 to 2220 Wp. These results suggest that, under the current regulation, PV system with lower capacity can deliver a larger portion of the generated electricity to user.

Carbon Reduction Benefit
Climate Transparency [20] estimated that approximately 761 gram of carbon dioxide are emitted to the air in order to generate 1 kWh of electricity in Indonesia.
Without using the PV system, the annual use of electricity at the Ontowiryo Mosque produces an emission of approximately 4017 kilograms of carbon dioxide. The Environmental Protection Agency [21] uses three sets of social cost estimates of carbon emissions with different discount levels, as can be seen in . This cost measures the negative impact of carbon emission on the environment from various externalities.  Based on the pricing above, the present value for the entire social cost of carbon emissions for 25 years, if the mosque does not use a PV system, is between 11.14 million and 50.51 million IDR. The use of PV system can reduce the amount of carbon emission and the social cost by 50% with using the partial supply PV system, and by 100% when the PV system completely supply the load demand.
When incorporated, the saving from emission social cost reduction can increase the financial attractiveness of the PV systems when compared to the carbon intensive fossil fueled power plants. It is important to note that the financial benefit of carbon emission reduction is not directly received by the PV system users under the currently implemented regulation. Nonetheless, it is also important to highlight the carbon emission reduction benefit as it helps the government to achieve the Nationally Determined Contribution goal and preserve the nature.

Financial Analysis
The results of the NPV analysis of the PV systems show negative numbers for all types of solar panels and both scenarios. This value varies with the lowest NPV at - Increasing the capacity of the PV system will reduce the use of electricity from the grid. Reduction of usage beyond the minimum usage limit incurs a negative effect on the annual revenue of the PV systems. The final electricity bill after using the PV system is equal to the minimum electricity bill for any usage lower than or equal to the minimum usage limit. In other words, at higher capacity of PV system, the amount of bill saving is capped at a maximum. On the other hand, the annual OPEX, which is a function of the capacity of the PV system, continues to increase.
Thus, the revenue, that is calculated by subtracting the annual OPEX from the annual bill saving, is decreasing at higher capacity of PV system. This condition explains why the PV system in the first scenario have lower annual revenue than those in the second scenario.

Abolishment of Monthly Minimum Usage
The financial feasibility study finds that the current regulation on the minimum usage limit for electricity from the grid incurs a detrimental effect on the PV systems' financial attractiveness at higher capacities. Therefore, in this scenario, the minimum monthly electricity usage from the grid is abolished while its impact on the PV systems' financial feasibility is studied. Such support policy has been tem- The impact of this regulation adjustment can be observed in figure 5. The REC355 TP2S 72 solar PV panel is taken as an example in this figure. It can be seen in this figure that the amount of annual revenue increases along with the increase in the number of panels and the capacity of the PV system before finally decreases after the netload is lower than or equal to the annual minimum usage limit, 2640 kWh.
The negative slope occurs as the annual bill saving is capped beyond this minimum limit, as explained in subsection 4.3, while the annual OPEX grows.
Meanwhile, in a scenario without the minimum usage limit, annual revenue increases along with the number of solar panels to the point where the net load is equal to zero. This change of regulation also increases the NPV of the PV in the first scenario for all types of the used solar PV panels. Figure 5 also shows in detail the increase of NPV at different number of PV panels employed in the PV system.
At higher capacity PV system, the increase of NPV is capped at around 35 million IDR, which occurs when the net load reaches zero.
The PV systems in the second scenario that only partially meet the load demand does not earn as much increase in their NPV as their net load is only slightly lower than or equal to the minimum usage limit. For PV systems in this design scenario, the increase of NPV results in the increase of capital expenditure recovery that ranges between 0.05% and 10%. Meanwhile for the complete supply design scenario, the increase of capital expenditure recovery is significantly higher as it ranges between 23.9% and 48.5%. However, despite the increase of NPV, the PV systems that is designed to completely supply the load demand still have negative NPV and lower than those that supply partially.

Rework of the Net-Metering Scheme
Currently only 65% of the exported electricity from the PV system will be compensated by the government. This decision is made by considering the transmission and distribution costs as well as the line losses of the grid system that accounts for 38% of the grid electricity tariff [22]. In this sensitivity analysis, the net metering scheme is reworked so that 100% of the exported energy from the PV system to the grid will be compensated.
The result shows that with this scenario, PV systems require smaller number of PV panels to either partially or completely meet the load demand of Ontowiryo Mosque, shown in figure 6. This condition leads to lower CAPEX, higher revenue and higher NPV. The NPV increase ranges between 19.5 and 40.8 million IDR for the PV system that completely supplies the load demand and between 5.7 and 11.9 million IDR for the PV system in the second modelling scenario that partially supplies the load demand. The increase of NPV also results in higher capital expenditure recovery that ranges between 20% and 40% for the complete supply scenario and between 15.3% and 21.4% for the partial supply scenario. Despite the increase, the NPV of all PV systems still yields negative values.

Higher Electricity Tariff
The non-subsidized tariff is taken from the average electricity generation basic cost provided in the PLN statistics 2019 and shown in figure 7 [19]. It is estimated that the non-subsidized tariff in 2021 and 2045 are at 1428 and 2205 IDR/kWh, respectively. The result shows that higher electricity tariff scales up the annual revenue.
The increase of NPV for all PV systems is capped at 19 million IDR due to the annual grid electricity minimum usage limit. Despite the increase, PV system that completely supplies the load demand still have negative values of NPV.  Figure 7 The average non-subsidized and subsidized electricity tariff in Indonesia extrapolated to 25 years. Adapted from [19]. On the other hand, when compared to other supportive measures, the implementation of higher electricity tariff results in the highest NPV increase for PV systems in the second scenario. The increase of NPV results in the increase of capital expenditure recovery that ranges between 26.5% and 59.%. As for the complete supply design scenario, the increase of capital expenditure recovery ranges between 13.6 and 28.4%.

Conclusion
In this paper, we have assessed the design of rooftop PV system on mosque and its economic implications. Our results show that the use of PV system allows a reduction of carbon emission as the power generation in Indonesia is dominated by the carbon-intensive power plants. This reduction also provides an indirect financial benefit that, when incorporated, can add leverages for the adoption of PV systems as an eco-friendly option to generate power.
Despite the carbon reduction benefit, at the current electricity tariff and capital costs of the components, and the implemented regulations around the adoption of PV system, it is financially unfeasible to install a rooftop PV system on a mosque.
The subsidized tariff of electricity results in lower annual electricity bill saving, which is the means of the PV system to recover its capital investment. The regulation on the minimum grid electricity usage limit prevents user to install high capacity PV systems due to the capped revenue and the increasing OPEX at higher capacity. This condition leads to our finding that PV systems with partial supply scenario have better financial attractiveness than the PV system that completely supply the load demand of the mosque. Intervention on the prevailing regulations such as rework on net-metering scheme and implementation of higher electricity tariff may increase the financial attractiveness of the PV systems.
Thus, the result of this paper should serve policy makers as a reflection for the prevailing regulation on the adoption of PV in the power sector. This paper provides an insight that the plan of increasing the share of renewable sources to 23% in the power generation by the year of 2025 can be hampered by the prevailing regulation. A more detailed exploration on the components of the PV system, especially those that are available locally, can enrich the study with a higher degree of context to the condition in Indonesia. Further analysis in this direction may lead to higher financial attractiveness of the PV systems as it cuts down the shipping cost in the procurement process. The need for further exploration is also supported by the large diversity of worship facilities from various beliefs in Indonesia. In some parts of the country, there are still many worship facilities with low socio-economic condition which require an eco-friendly means to be independent in meeting their energy needs.