Designing and Analysing a PV/Battery System via New Resilience Indicators

Abstract

The increasing frequency of natural disasters in refugee camps has highlighted the urgent need for a dependable power source. In humanitarian camps, a reliable power supply is essential for meeting the basic daily needs of the residents. However, the conventional power systems in these camps often experience interruptions. To address this issue, microgrids have emerged as a viable solution. Although several studies have recognized the resilience benefits of microgrids, their application in refugee camps has been relatively limited. Hence, this study presents a grid-tied microgrid that combines photovoltaic and battery systems, designed using REopt lite web version software, to fulfill the energy requirements of Rohingya refugees in a selected camp located in Ukhia, Cox’s Bazar. Additionally, this study introduces four new indicators of resilience to evaluate the performance of the designed system. The findings reveal that the proposed microgrid consists of 5685 kW of photovoltaic capacity and 9011 kWh of battery capacity, enabling it to handle a 15 h power outage and resulting in substantial life-cycle savings of USD 2,956,737. The payback period for this resilient system is calculated to be 6.37 years, with an internal rate of return (IRR) of 12.2%. Furthermore, the system emits fewer emissions compared to other analysed modes in resilient operation, emphasizing its sustainability. In conclusion, the designed microgrid successfully enhances the reliability of the power supply in refugee camps.

1. Introduction

A reliable and resilient power supply is necessary to ensure safety and uphold essential human rights these days, though it is limited in many parts of the world [1,2,3,4]. Developing and under-developed countries often face inadequate and non-existent power-grid infrastructure [5,6]. In recent years, microgrid technology has appeared as a promising technology to overcome this challenge, delivering a flexible and decentralized energy generation and distribution approach [7,8,9]. Microgrids are self-contained energy systems that can work independent of the larger power grid, supplying reliable and cheap power to underprivileged neighbourhoods. Various renewable energy sources like wind and solar can be used to power microgrids [10], and they can also be designed to meet the specific needs of the residents [11,12,13]. Power electronics devices are also used in microgrids to convert, control, and maintain the power flow between different sources and loads [14,15,16,17]. Integrating automation with power electronics has made microgrids more resilient than the conventional grid since they supply electricity during grid disruption [18,19].

Refugee camps are in remote areas without reliable grid connections or power supplies [20]. This significantly impacts refugees since, with proper and reliable electricity, refugees will be provided with education, healthcare, and security [21]. Microgrids can come in handy in this regard. Several studies have designed microgrids to supply electricity to remote areas. Chowdhury et al. designed a PV battery microgrid for Rohingya refugees and discovered the system’s resilience benefits. This study reported that the system could withstand a 24 h grid outage [22]. Similarly, a PV/battery/ CHP system can handle a 24 h grid outage in Texas. The system consists of 3933 kW of PV, 208 kW CHP, and 4441 kWh of storage, and the system has a net benefit of USD 1,007,204 over the project’s life span [23]. Chowdhury et al. also evaluated the sustainability of a PV/generator/wind/battery system for a Rohingya community. The system can supply electricity at the cost of 0.35 USD/kWh [24]. Masrur et al. proposed a microgrid for a critical facility of an airport and determined that the designed microgrid could save more than seventy-three thousand dollars and withstand 718 h of a grid outage [25]. Marqusee et al. applied an REopt-based optimization model to study the resilience and financial benefits of microgrids combined with PV, battery, and diesel generators, and found that hybrid microgrids are more resilient and contribute to more financial savings than traditional islanded systems [26]. Farthing et al. designed a novel solar plus storage system to optimize microgrids’ health, climate, resilience, and energy savings for 14 US cities [27]. The study found that monetizing and the additional optimizing of health and climate benefits enhance the net present value of microgrids in comparison to resilience and energy saving bills. Anderson et al. incorporated duration-dependent “customer damage functions (CDFs)” into microgrid-based modelling and found that understanding the value of lost load gives opportunities to minimize the life cycle cost of energy [28]. Becker et al. performed a techno-economic analysis of a geothermal energy-based energy system for a cold-climate “zero energy community” and found that the application of a geothermal-based energy system in the place of PV will significantly minimize energy import and export [29]. Nguyen and Mitra determined the impact of adding gas turbines on energy systems’ frequency stability. They found that gas turbines with a fast response ability positively impacted the recovery period of system frequency [30]. Mitra and Nguyen introduced grid-scale “virtual energy storage” that stores surplus energy to enhance renewable energy generation before the frequency stability of the system is compromised [31]. Pandit et al. optimized the size and number of batteries in the “battery exchange-based electric vehicles” (BEVs) and the microgrid to minimize the overall system cost and improve the performance of the microgrid. The optimization results show that the proposed approach can reduce the overall system cost and improve the microgrid’s performance. In particular, the research demonstrated that using BEVs with exchangeable batteries can significantly reduce the battery bank size required in the microgrid, leading to cost savings [32]. Muhtadi et al. conducted a case study on a remote island in Indonesia to evaluate the technical and economic feasibility of implementing an entirely renewable-based energy system to meet the island’s energy demand [33]. They developed a model to determine the optimal sizing and configuration of the system based on the island’s energy demand, renewable energy resources, and storage requirements. The authors concluded that the proposed renewable-based energy system could be a viable option for remote islands and off-grid communities to achieve energy independence while reducing their reliance on fossil fuels and greenhouse-gas emissions. Similarly, microgrids’ resilience benefits were also observed for irrigation [34], water infrastructure [35], hotels [36], health care [22], and buildings [37].

The literature has effectively established the advantages of microgrids in terms of resilience. However, the benefits of microgrids in refugee camps are still a recent topic. This study uses the REopt lite web version software to create a microgrid model consisting of a PV and battery system to assess its resilience in a Rohingya refugee camp in Bangladesh. The refugee population showed a significant increase from 2015 to 2017, and currently, over one million Rohingya population live in Cox’s Bazar camps in Bangladesh, suffering from energy-related issues [38]. A distributed energy system can play an essential role in this regard. This analysis selects a microgrid to supply electricity to a hypothetical hospital. For this, a camp in Ukhia, Cox’s Bazar, is selected.

A camp able to accommodate a population of 24,026 was considered [21]. The hospital has a floor area of 241,351 FT². The hospital has five floors. It is assumed that one thousand two hundred daily patients will be outpatients. The annual energy consumption of the hospital is 8,156,140 kWh [39]. The critical load of the hospital is assumed to be 60% to give the system a safety margin. This load indicates that the critical load can still be appropriately met even if there are fluctuations in energy generation or supply changes. It acts as a buffer to retain the system’s dependability and stability. Critical or essential load is the type of load that must always be met. A hospital has several critical loads, such as air conditioning, refrigerators, and lighting. The energy consumption of the hospital can be found in Figure 1. Figure 1 shows that energy consumption is highest during August (summer season). More loads will be imposed on the system since there will be more need for water and air conditioning. Because of that, it is considered that the outage will occur in August to observe the system performance. We considered a 15 h outage because there is a 17% possibility of 12–24 h outage occurrence due to natural disasters [40].

Microgrid modelling was proposed in this study to resolve the outage issue. The economic cost and environmental impacts of three modes (business as usual, financial, and resilient) were evaluated. This study introduces four new resilience indicators, which provide an innovative and essential aspect to assessing the system’s performance. The aim is to analyse the designed microgrid via indicators. Apart from this, different tariff (peak rate) rate was considered in our study. The authors anticipate that this study will make a valuable contribution to sustainable energy and humanitarian aid, as it provides helpful recommendations to local and international communities.

2. Methodology

2.1. Optimization Simulation by REopt

Different studies have used a set of different algorithms for multi-objective optimization [41,42,43]. A mathematical optimization technique called MILP optimization was used in this study, specifically REopt Lite, to evaluate the optimum sizing of a distributed energy system. Modelling an energy system often requires making tactful decisions, such as selecting equipment sizes or figuring out how many units to install. MILP can address these careful decisions and discover optimal or near-optimal solutions regarding the options. The process of REopt involves an optimization module (OM) and a simulation module (SM). OM prioritizes minimizing a location’s life-cycle energy costs (LCC). LCC is a statistical analysis for this study that considers all expenses related to a project option, including taxes and incentives. OM also ensures that the designed microgrid can supply power to critical or essential loads without the grid during a defined or pre-defined blackout. OM generally includes financial and resilience analysis. The dispatch strategy and size optimization are performed in financial analysis to reduce LCC. In contrast, the same thing is also performed in resilience analysis. However, there is an additional requirement that the system provides power to the essential load during the outage period without assistance from the utility grid. The SM takes the recommended technologies and sizes from the OM as inputs and evaluates the yearly resilience output of the system by simulating outages every hour of the year (8760 times). The main disparity between SM and OM is the modelling techniques used. The OM fixes the system size, dispatch strategy, and outage period, while the SM takes in system size and simulates outages every hour of the year instead of a single outage. A load-following approach is used to determine hourly dispatch in each outage simulation. The optimization simulations were evaluated using Equations (1)–(8) [44].

$$minLCC=\min \left(C_{Eg}+C_{OM}+C_{Dn}+C_{PVBAT}\right)$$

(1)

$$C_{Eg}={\displaystyle \sum }_{l\in \mathcal{L},hϵ\mathcal{H}}(F_{tlh}^{pd}\ast P_{tlh}\ast c_h^e)$$

(2)

$$C_{PVBAT}={\displaystyle \sum }_{t\in \mathcal{T}}(X_t\ast c_t)+(B^{kWh}\ast c_{kWh}^b)+(B^{kW}\ast c_{kW}^b)$$

(3)

$$C_{Dn}={\displaystyle \sum }_{r\in \Re }(d_r\ast c_r^d)+{\displaystyle \sum }_{m\in \mathcal{M}}(d_m\ast c_m^d)$$

(4)

$$C_{OM}={\displaystyle \sum }_{t\in \mathcal{T}}(X_t\ast c_t^{OM})$$

(5)

These three load constraint expressions are used to reduce LCC.

$${\displaystyle \sum }_{t\in T}(F_{t{l}^{Ex}h}^{pd}\ast P_{t{l}^{Ex}h}\ast F_t^{dgr})\le L_{lh}, \forall h ϵ \mathcal{H}$$

(6)

$${\displaystyle \sum }_{t\in T}{F}_{t{l}^{R}h}^{pd}\ast P_{t{l}^{Ex}h}\ast F_t^{dgr}+B_h^{-}\le L_{lh}, \forall h ϵ \mathcal{H}$$

(7)

$${\displaystyle \sum }_{h\in \mathcal{H}}{F}_{PVlh}^{pd}\ast P_{PVlh}\ast F_{PV}^{dgr}\le {\displaystyle \sum }_{h\in \mathcal{G}}{L}_{l^{S}h},\forall l ϵ \mathcal{L}$$

(8)

Equation (6) confirms that the total energy generated locally never exceeds all energy sources’ combined load. A combination of PV, grid, and battery power must always meet the site load. In contrast, according to Equation (7), Equation (8) places a cap on the amount of electricity that can be produced by solar energy so that it is equal to or less than the load of the hospital. Equation (9) ensures that the power produced by solar energy is equivalent to the system size selected for each stage across all loads. This is the load constraint used in the model.

$${\displaystyle \sum }_{l\in \mathcal{L}}{P}_{tlh}\le X_{th}, \forall h ϵ \mathcal{H}$$

(9)

Equations (10)–(13) depict the constraints of battery-storage technology. This includes charging, discharging, degradation, and the battery’s charge level over a specific period. When the value of $Z_h^{B+}$ is 1, then it represents the battery charging state. The battery discharging state is represented by the value of $Z_h^{B-}$ and its value is 1.

$$B_h^{+}={\displaystyle \sum }_{t\in T}(F_{t{l}^{B}h}^{pd}\ast X_t\ast F_t^{dgr}\ast ɳ{ }_B),\forall h ϵ \mathcal{H}$$

(10)

$$B_h^{SOC}=B_{h-1}^{SOC}+B_h^{+}-B_h^{-},\forall h ϵ \mathcal{H}$$

(11)

$$B_h^{-}\le B_{h-1}^{SOC}, \forall h ϵ \mathcal{H}$$

(12)

$$Z_h^{B+}+Z_h^{B-}\le 1,\forall h ϵ \mathcal{H}$$

(13)

Equations (14) and (15) show demand rate constraints. The demand should be greater than or equal to the monthly grid electricity.

$${\displaystyle \sum }_{h\in {\mathcal{H}}_r, l\in \mathcal{L}}{P}_{Glh}\le d_{r,} \forall r ϵ \mathcal{R}$$

(14)

$${\displaystyle \sum }_{h\in {\mathcal{H}}_m, l\in \mathcal{L}}{P}_{Glh}\le d_{m,}\forall m ϵ \mathcal{M}$$

(15)

Equation (16) indicates that PV technology’s size is equal to or less than the net metering level if functioning at that level, and zero otherwise.

$${\displaystyle \sum }_{t\in T}{X}_t\le L_{sv}^{NEM}\ast Y_{sv}, \forall v ϵ V, s\in S$$

(16)

Some of the applicable restrictions listed in [44] were considered in this study. This research did not consider other technology constraints, except PV and battery. From this model, the benefits of the microgrid’s resilience during the period of grid de-energization can be clearly understood, in addition to its reduced cost and better energy consumption. Also, actual critical loads might be entered into the model instead of percentages.

The Net Present Value (NPV), Payback period (PB), and IRR (Internal rate of return) are the economical parameters of financial analysis in REopt software. NPV is the value of all future cash flows with a discount rate in today’s currency, while IRR represents the discount rate that causes the NPV to be zero. PB is a metric that shows the time needed to recover the investment for one project. The NPV, PB, and IRR were determined by Equations (17)–(19), respectively.

$$NPV={{\displaystyle \sum }}_{z=0}^N\frac{c_{1,z}}{{\left(1+r\right)}^z}$$

(17)

$$0={{\displaystyle \sum }}_{z=0}^N\frac{C_z}{{\left(1+IRR\right)}^z}$$

(18)

$$PB=\frac{I}{P}$$

(19)

2.2. Grid Simulation

The nearby utility grid is a dependable power source for the refugee camp and does not require maintenance, capital, or operating expenses. The tariff for the utility rate only covers the costs related to the energy supplied by the grid. Equation (20) was used to measure the electricity provided by the grid, and is given below [23]:

$$P_g\left(t\right)=P_l\left(t\right)-\sum (P_{PV,}{P}_{batt})$$

(20)

It is essential to include the rate at which electricity is charged to model the electricity system. A peak rate is used for this study, as seen in

Table 1. Schedule of the electricity rates of Bangladesh’s national grid [45].

Model Rate (for 11 kV Line)	Grid Power Price (USD/kWh)	Demand Rate (USD/kW/month)	Sell Back Rate (USD/kWh)
Off-peak	0.085	0.577	0.050
Flat rate	0.094	0.577	0.060
Peak	0.120	0.577	0.080

.

2.3. Simulation for Photovoltaic Module

The conversion of the energy available in solar resources into electricity is possible using a photovoltaic module [46,47]. Equation (21) was used to calculate the power available from the solar resource [22]:

$$P_{PV}=C_{P_{PV}}{D}_{PV}\left(\frac{I_r}{I_{r_{STC}}}\right)\left[1+{\mathsf{\alpha }}_{\mathrm{P}}\left(T_C-T_{C,STC}\right)\right]$$

(21)

Equation (22) provides a way to measure how effectively the PV module can convert solar energy into electrical power, considering maximum power and standard test conditions [22].

$${\mathsf{\eta }}_{STC}=\frac{C_{P_{PV}}}{A_{PV}{I}_{r_{STC}}}$$

(22)

In this analysis, the capital cost of the solar panel system was estimated to be USD 310 per kilowatt (kW), and the system was expected to have a lifetime of 25 years. The operational and maintenance costs were assumed to be USD 17 per kW per year [22], and the system was planned to be installed on the ground rather than on the hospital’s roof without any tracking arrangement. Additionally, surplus energy generated by the PV system beyond the load demand and battery charging requirements was not utilized in this study and, instead, wasted. The National Renewable Energy Laboratory’s (NREL) PVWatts application was used in conjunction with the REopt software to estimate the electricity production of the installed PV system. It was assumed that for every 2.42811 hectares of available space, one MW-DC (megawatt direct current) of PV would be installed, with a DC-to-AC size ratio of 1.2 and a system loss of 14% [22].

2.4. Simulation for Battery

A battery system is required to store excess energy generated by solar panels to account for the intermittent nature of renewable energy sources. This stored energy can provide power during periods when solar resource is unavailable. The size of the critical battery system can be calculated using Equation (23) [22]:

$$C_{a{p}_{bat}}=\frac{E_{load}DA}{{\mathsf{\eta }}_{con}{n}_{batt}DOD}$$

(23)

Lithium (Li)-ion batteries have gained significant attention among consumers due to several advantages [48,49,50]. This analysis assumes a Li-battery to be implemented because of its high energy density and longer life span [51]. A li-battery with a life expectancy of 10 years was considered here. The capital cost of the battery system was estimated to be USD 419 per kW, and the power capacity cost of the battery was USD 775 per kW [22,23]. It was also assumed that the grid could charge the battery as needed. The battery was initialized with a state of charge (SoC) of 50%, and the minimum SoC was set to 20%. It was also assumed that the inverter would be replaced once during the system’s lifetime, and the replacement cost was included in the annual operational and maintenance costs. The efficiency of the inverter was assumed to be 96%. The whole system can be found in Figure 2.

2.5. New Resilience Indicators

2.5.1. Energy Flexibility

This metric assesses the system’s capacity to modify its energy demand and supply in response to environmental factors. It can be determined by examining the system’s ability to adapt to real-time power output, battery storage, and energy efficiency adjustments. Energy flexibility can be found in Equation (24):

$$\mathrm{Energy} \mathrm{flexibility}=\frac{Energy demand and supply adaptation during an outage}{Total possible energy demand and supply adaptation}\times 100$$

(24)

2.5.2. Economic Resilience

This indicator estimates the system’s capability to remain profitable in disruptions. It can be calculated by looking at the system’s cost-effectiveness, potential for income generation, and capacity for market-change adaptation. Equation (25) can be used to find economic resilience.

$$Economic Resilience=\frac{\left(Attainment of monetary goals during disturbances\right) }{Monetary goals}\times 100$$

(25)

2.5.3. Environmental Resilience

The system’s capacity to maintain ecological sustainability in the face of disturbances is measured by this indicator. It can be calculated by studying the system’s ecological impact, including greenhouse gas emissions, waste generation, and resource depletion. Environmental resilience can be found in Equation (26):

$$Environmental resilience=\frac{Environmental performance during outage }{Environmental performance under normal operating condition }\times 100$$

(26)

2.5.4. Community Resilience

This indicator determines the system’s capacity to meet the demands of the neighbourhood community in the event of disruptions. It can be calculated by looking at the system’s capacity to offer vital services to the neighbourhood during emergencies, including emergency electricity. Community resilience can be found in Equation (27):

$$Community Resilience=\frac{Electricity provided during disruption}{Total energy demand}\times 100$$

(27)

3. Results and Discussion

This study examined three scenarios: business as usual (BaU), financial, and resilience. The BaU scenario depicts that only the grid will supply electricity to the load. In the financial scenario (FS), the sizes of wind, PV, and battery systems are optimized along with a battery dispatch strategy to minimize the life-cycle cost of energy (LCOE). In the resilience scenario (RS), the same technologies are optimized, along with backup generators, to ensure that critical loads can be sustained during a 15 h grid outage while keeping LCOE low. The simulation results show that the system in the RS, which consists of 5685 kW of PV and 9011 kWh of battery, is better able to sustain the simulated outage than the FS, which has 3652 kW of PV and cannot support the outage. Similarly, a 10 h outage is modelled in the system keeping all the inputs constant for sensitivity analysis to fulfill the load. The optimum system comprises 4817 kW of PV and 5362 kWh of battery. This range proves the energy flexibility of the microgrid.

The NPV of the FS system for the 15 h outage system is 22% higher than the NPV of the RS system presented in

Table 2. Comparison between BaU, resilience, and financial scenario.

Parameters	Business as Usual (BaU)	Resilience	Financial
Average Annual PV Energy Production		7859 MWh	5049 MWh
Average Annual Energy Supplied from Grid	8140 MWh	2232 MWh	4688 MWh
Total CO2 Emissions in Year 1	4947 tons	1357 tons	2849 tons
Year 1 Utility Electricity Cost—Before Tax
Utility Energy Cost	USD 976,740	USD 267,817	USD 562,526
Utility Demand Cost	USD 9561	USD 4544	USD 9146
Total Life-Cycle Costs	USD 11,809,978	USD 8,853,241	USD 8,217,548
Payback Period	N/A	6.37 years	2.1 years
Internal Rate of Return	N/A	12.2%	38.8%

. NPV refers to the total cost savings that a project will generate over its entire duration, considering the time value of money. The payback period in the RS system is almost seven years, while in the financial scenario, it is only two years. This result indicates the investment will take significantly longer to pay back its initial RS system cost than the financial scenario. But the RS system can endure the outage while the FS system cannot. In addition, for a 10 h outage system, NPV can be observed to be USD 3,268,978. This outcome proves the economic resilience of the proposed microgrid.

In the BaU scenario (for a 15 h outage system), the grid supplied 8,139,504 kWh of electricity to fulfill the demand of the refugee people. The annual PV production in the RS is 56% higher than in the FS. A proportion of 73% of electricity comes from renewable energy in the RS, which is 30% more than in the FS. The system is also more environmentally friendly in the RS than the FS. The total CO₂ emissions in the resilience scenario are 52% lower than in the financial scenario. This output demonstrates that the microgrid can withstand environmental challenges and recover quickly from disruptions.

A 15 h outage was assigned to observe the system’s performance. The outage was assumed to occur at 2 PM on August 9. The system’s performance during the grid outage can be found in Figure 3. During the blackout, the grid stopped supplying electricity to the load. PV electricity was provided to the load during the outage till 5 PM. PV cannot generate electricity due to the absence of a solar resource. Now, the battery discharged till 5 AM to meet up the load. The microgrid successfully maintained power supply to critical loads during the 15 h outage. The battery storage system provided backup power during the outage, preventing the need for a costly and potentially unreliable backup generator. The PV system could recharge the battery during daylight hours, allowing the microgrid to continue operating without grid power to supply electricity to the hospital, presenting the community resilience of the microgrid.

The microgrid’s overall reliability was improved by adding energy storage, reducing the impact of power outages, and smoothing out fluctuations in PV output. The probability of surviving the outage can be found in Figure 4, which is a direct result of the improved reliability of the microgrid system (please check supplemental file for the procedure). In addition, if the cost of power outages is considered, estimated at USD 100 per kilowatt-hour, implementing the microgrid system would result in an avoided outage cost of approximately USD 23 million over the project lifetime. This represents the monetary advantage attained by decreasing outages after implementing the microgrid. Power outages can have severe repercussions in a regular grid system, specifically in vital areas like refugee camps. These outages can negatively impact essential services, healthcare facilities, communication networks, and camp residents’ general health and safety. This incident would result in a substantial financial loss. By implementing the PV/battery system, a total cost saving of around USD 24 million throughout the project’s life cycle can be gained in the camp. In addition, the payback period of the proposed system is less than indicated in the study conducted by Chowdhury et al. [22] which suggests a more immediate return on investment and a quicker retrieval of capital. Moreover, the LCOE of the proposed system is USD 0.020/kWh, which is 94% less than the outcome observed in [22].

Project Validation and Future Recommendations

Implementing PV/battery/diesel microgrids for critical facilities can save USD 440,191 over the project’s duration [52]. Similarly, implementing PV and batteries in a commercial building could save USD 50,000 throughout its lifespan [36]. In addition, a solar/battery microgrid in an office building could increase power resilience during a 4 h outage and save up to USD 112,410 throughout the project [53]. Based on our analysis, several recommendations are provided to ensure the successful implementation of microgrids in refugee camps:

• Microgrids in refugee camps should be designed, installed, and maintained by governments and humanitarian groups with adequate funding. The standardization of microgrid technology and components should be encouraged to guarantee the interoperability, dependability, and affordability of microgrid systems.
• Governments should provide legal frameworks that support the installation of microgrids in displaced camps. These guidelines should support the creation of creative finance structures and encourage participation from the private sector.
• Governments and humanitarian groups should offer financial incentives to encourage the use of microgrids. For instance, businesses that engage in microgrid projects in refugee camps might receive tax benefits.
• Governments and humanitarian groups should offer technical support to microgrid developers to guarantee the successful installation and operation of microgrids in refugee camps. Power electronic devices can often induce harmonics in the DC voltage, which will degrade the DC voltage quality [54]. Frequent charging and discharging can also cause harmonics in the system [55]. So, careful consideration needs to be made in this regard.
• Capacity-building programs should be established to train local personnel on the maintenance and operation of microgrid systems. These steps will ensure the sustainability of the systems over the long term. The microgrid is often subjected to theft in certain regions where electricity theft is prevalent [56]. Proper techniques should be used to monitor this situation [57].
• Cooperation between governments, humanitarian organizations, and private sector actors should be encouraged to facilitate the successful installation of microgrids in refugee camps. This measure will make the energy system more robust and able to endure disturbances such as natural disasters.

4. Conclusions

The performance of a PV/battery microgrid was modelled in this analysis to observe the resilience benefits. Four new resilience indicators, namely energy flexibility, economic, environmental, and community resilience, were proposed to evaluate the microgrid performance. Different outage periods were modelled to see if the microgrid could survive. It was found that the microgrid can withstand the outage, resulting in life-cycle savings of USD 2,956,737. The system comprises 5685 kW of PV and 9011 kWh of battery. This means that financial gains were made possible by implementing the microgrid system, in terms of cost reductions and avoided expenditures throughout the project’s anticipated lifespan compared to the grid-only system. A smaller payback period and higher IRR make the simulated system attractive for Rohingya refugees. The resilient system emits less CO₂ than any other system. For future investigation, the thermal energy aspect of the hospital could be considered, and for this, PV/CHP/battery system could be modelled. This study has simulated resilience benefits for hospitals, so for future studies, other critical infrastructures like data centres and police departments could be considered. Also, the system was modelled in the present research with a pre-defined outage, which is one of its limitations. Applying a stochastic technique to predict the outage time and length in the optimization module could be considered for future research purposes.