Techno-Economic Feasibility of a Grid-Connected Hybrid Renewable Energy System for a School in North-West Indonesia

Background: Schools typically have high diurnal fluctuation in electricity demand, with peak loads during daylight hours, which could be adequately met through harnessing solar renewable resources. This study demonstrates the strength of techno-economic assessment in selection and optimization of a grid-connected hybrid renewable energy system (HRES), utilizing local renewable resources to fulfil the daytime electricity demand for a school in northwest Indonesia. Methods: Three different scenarios are developed for optimizing the HRES configurations, comprising of PV panels, Wind turbine, Battery and Inverter. The following optimization parameters are used—one, technological performance of the HRES, in terms of their energy output to fulfil the energy deficit; two, economic performance of the HRES, in terms of their net present cost (NPC) and payback periods. Results: A clear trade-off is noted between the level of complexity of the three HRES, their renewable electricity generation potentials, NPC and payback periods. Scenario II, comprising of Solar PV and Inverter only, is found to be the most feasible and cost-effective HRES, with the optimized configuration of 245 kW PV capacity and 184 kW inverter having the lowest initial capital cost of US$ 51,686 and a payback time of 4 years to meet the school’s annual electricity load of 114,654 kWh. Its NPC is US$ −138,017 at the 20th year of installation. The negative value in year 20 is achieved through the sale of 40% of the renewable energy back to the grid. Conclusions: Techno-economic assessment can provide useful decision support in designing HRES relying on solar energy to serve predominantly daytime school electricity requirements in tropical countries.


INTRODUCTION
The Indonesian government has set target to sustainably harness energy by increasing the share of new and renewable resources in primary energy supply to reach 23% by 2025 and 31% by 2050 [1,2].
However, while the last decade has seen rapid growth in renewable energy generation in many parts of the world [3], the grid supply in Indonesia, and more widely in South-east Asia, still heavily relies on conventional fossil-based electricity generation [1,4]. The majority of the Indonesian people are still not fully aware of the bountiful solar resources which can be harnessed for sustainable energy generation [5]. This is contradictory to the fact that increased heat stress among the local population has reduced the Indonesian annual GDP by 6%, attributed mainly to physiological impacts and reduced labor productivity owing to its close proximity to the Equator [6].
The statistical data of the state-owned electricity company of Indonesia shows that in 2016 only 12.16% of the overall electricity was generated from renewable energy sources, of which the majority was from hydro and geothermal energy [7]. Solar and wind power in total only contribute to mere 0.02% of the entire electricity generation, despite having an average daily solar radiation of 4.80 kWh/m 2 [8]. However, only 0.038% of this resource is converted into electrical energy at the moment. The typical average wind speed in the country ranges between 3 and 6 m/s, however, only 0.005% of this resource was used into the national energy mix in 2016 ( a school unit in Turkey showed that the electricity generation solely from renewables was not sufficient to compensate the overall energy demand of the school [13]. Off-grid application of solar and wind power generation with diesel generator and battery backup was found to fully meet the energy demand of a rural school in Morocco at an optimized energy cost of about 1.12 $/kWh [14]. Analysis of a grid-connected photovoltaic system for a school in Wellington, New Zealand, reported that the power generated from the system escalated only during the summer time [15].
Another study mainly focused on optimization of several HRES components to compensate for the heating and electricity demand of a school building in South Korea [12]. A study assessing the viability of HRES in a school building in Greece found that the installation could provide mitigation of climate change, resources protection, costs reduction, and energy saving [16].

Site description
A school building in the city center of Banda Aceh in the Aceh province in northwest tip of Indonesia is selected as the case study, located 7 m above sea level at 5.55° N and 95.35° E (Figure 1). The site was hit by the devastating tsunami in 2004; over 90% of the region neighboring the school is electrified, but largely supplied with a mix of diesel and hydro power plants, with approximately 98% of electricity originating from fossil fuel combustion [7]. However, the region has approximately 4.1 kWh/m 2 average daily solar radiation [2], providing huge potential for a gridconnected renewable energy system.

Electricity demand profile
The daily electricity consumption profile is generated using the real

Renewable resource availability
Most of the previous studies have used freely accessible research databases on renewable resources from their respective governmental or national weather databases [12,16,22,23]; the National Institute of Water Database, has been used to generate the monthly averaged solar irradiance profile for the city of Banda Aceh (based on latitude and longitude) (Figure 4) [24]. GHI is considered appropriate for estimating PV power outputs as it is the sum of beam radiation (also called direct normal irradiance or DNI), diffuse irradiance, and ground-reflected radiation.
Based on the data, the location is found to have a fairly constant solar irradiance profile throughout the year, which is owing to its location near the equator. However, there is slight decrease in the magnitude of solar irradiance in the last 3 months of the year, attributed to the onset of the rainy period over these months. Wind speed: The monthly average wind speed profile at the school location has been estimated using the hourly wind speed data obtained from the NREL database for the city of Banda Aceh ( Figure 5).

Energy system component specification and pricing
The hardware spec for the different HRES components used in this study and their initial capital costs are shown in Table 2. The lifespan of the energy generating hardware components (PV and wind turbine) is assumed to be 20 years that of the inverter as 15 years and the battery life is assumed to be 10 years. The selected PV module has maximum power of 250 W, efficiency of 15% and the annual operational & maintenance cost of US$ 100 [25]. A relatively smaller size single wind turbine is considered for installation at the school with the maximum power output of 10 kW, rotor diameter and tower height of 5.5 m and 15 m, respectively [26]. All the solar panels have been assumed to be installed on building rooftops facing due north and tilted at 5° to the horizontal to allow for draining of rain water. The key consideration in determining appropriate rooftop location for citing the solar panels is their non-obstruction from shadows throughout the daylight period. It was considered that only one wind turbine could be installed, given the small fetch offered by the builtup space in the school premises.

Energy pricing
The Feed-in-tariff introduced by the Minister of Energy and Mineral Resources (MEMR) vide regulation No. 50/2017 on power purchase for a range of renewable energy generator is applied to this study. Accordingly, the selling price of electricity from the grid has been set to the national average production cost in Indonesia (US$ 0.12 per kWh) [7]. Therefore, the price of buying electricity from the grid is set to be equal to the selling price. In other words, the excess power generated by HRES in the school is assumed to be sold to the grid at the same unit price of US$ 0.12 per kWh.

Modelling Methods
The Hybrid Optimization of Multiple Energy Resources (HOMER Pro ® ) model, developed by the National Renewable Energy Laboratory [29], is used to assess the techno-economic feasibility of the proposed HRES to be installed at the school. The modelling method followed the suggested approach in the literature [22,30,31], requiring several input parameters in terms of the wind and solar resources, energy demand, system components, and the specified cost of each component.  compensated by the power stored in the battery when the grid is unable to provide the electricity. Moreover, the excess energy from the renewables can also be sold back to the grid when the battery is fully charged. This is specifically suitable for utilizing the battery backup in meeting the unmet loads. The drawback for this configuration is that the total cost is expected to be relatively high.

Scenario Analysis
Each of the three modelled scenarios present unique attributes in terms of their optimum configuration size of the renewable energy technologies affecting their feasibility and also the total cost of the system over their 20-year operational life. The priority of this exercise is to find the configuration with the lowest total cost and the fastest payback period. year there will be an income of that amount from the sale of electricity back into the grid. Additionally, the payback time of this design was estimated to be 9 years. In this configuration, 88.4% of the school electricity demand is estimated to be compensated by the power from the PV. Additionally, the power from WT and the grid contributed only 7.81% and 3.77% of the load respectively. The simulation showed that the power produced by WT is inferior to that by PV, which is acceptable for the location, based on the availability of low wind speeds. This configuration is advantageous to be deployed in the school setting as it appears to meet the electricity requirements for the school during the peak operation time. Figure 7a shows that most of the load can be compensated by the renewable power (upper line) generated from the optimal configuration for this scenario (only 2.82% of the generated power were consumed by the school and the rest 97.2% were sold back to the grid). This corroborates the findings reported in a previous study [12], which stated that the RES installation at a school is suitable in accordance of its energy demand profile and condition of the location.
Scenario II-This scenario maximizes the potential for solar generation while reduces the capital expenditure by removing the deployment of wind turbine and battery backup. However, as expected, the HRES power output of this scenario is relatively lower than the previous scenario and restricted to daylight hours (meeting 94.5% of the electricity demand, with the 5.50% fulfilled via grid purchases). Figure 7b shows the typical weekly energy demand profile of the school and the total electricity supply from the optimal configuration for this scenario. On an average, the school consumed around 30% of the total power generated and the remaining 70% was sold back to the grid. However, as in case of Scenario I, some portion of the energy requirement during early morning hours could not be met from the renewable generation when the solar irradiation is not quite intense. Hence, maintaining grid connectivity was considered essential to avoid any loss of supply during these hours. The outcome of the estimation showed that the amount of energy which could be sold back to the grid was directly proportional to the inverter output power.
Scenario III-In this scenario the deficit power could be either drawn from the batteries or bought from the grid. Similar to the previous two scenarios, the power from renewables could not fulfil the electricity demand of the school in the early morning hours. However, the reliance on the grid supply was compensated by drawing the deficit power from the battery storage, leading to minimal grid electricity purchase for this scenario. Figure 7c shows how the battery system accommodates the morning time load, indicated by dip in the Battery energy (top black line).
The use of battery was observed during the week-days and Saturdays.
However, due to the low electricity demand on Sundays, the batteries were noted to be fully charged and the excess power produced by the renewables were sold to the grid.

Cost efficiency
The initial capital costs, NPC values and payback periods for the optimized HRES configurations to meet the annual electrical load of the school of 114,654 kWh for the three scenarios, and the corresponding annual energy flows to the grid (purchase and sell), are shown in Table 3.  It is noteworthy, the study mainly focused on renewable energy provision to meet the existing school energy demands, without considering further modification of the building design and passive cooling strategies, e.g., HVAC efficiency enhancement and/or cooling capacity enhancement through inclusion of innovative PV-thermal technologies, which are still undergoing R&D improvements and thereby lacking cost-competitiveness [21].

CONCLUSIONS
This study demonstrates the strength of techno-economic assessment as a decision support in developing hybrid renewable energy system

CONFLICTS OF INTEREST
The authors declare that there is no conflict of interest.

ACKNOWLEDGMENTS
The first author acknowledges the support of the government of Aceh Province, northwest Indonesia in conducting this research.