Optimal Sizing and Operation of PV–Diesel–Battery Hybrid Energy Systems for Electric Vehicle Charging Using Homer Grid

in [1] investigates sustainable hybrid configurations for EV charging stations , emphasizing how local conditions drive optimal sizing and performance. Despite strong evidence on site specificity, the analysis does not deeply couple stochastic EV arrival patterns with multi- constraint dispatch to co-optimize cost, reliability, and emissions across realistic tariff regimes. Also, the study in [2] compares battery technologies in grid-connected PV with EV contexts, quantifying techno-economic trade-offs and storage impacts on system performance. However, the battery-chemistry focus leaves broader hybrid sizing (PV, diesel, inverter limits) and detailed reliability constraints (SOC bounds/LPSP≈0 for fast charging) only partially explored. In [3], PV-powered charging under varied Indian climate zones was optimizes, showing strong climatic sensitivity of optimal designs. But the climate-centric optimization does not fully address generator rightsizing or diesel–storage interplay for resilience where grid reliability is weak or demand charges are high. In addition, reference [4] studies grid-tied hybrid renewable systems for EV charging, highlighting how PV with storage can cut fuel use and costs compared to conventional setups. While grid-tied economics are assessed, there is limited treatment of off- grid/weak-grid regimes and the operational constraints needed to guarantee near-zero unmet load in DC fast-charging contexts. A broad review of renewable-integrated EVCS covers siting, power quality, storage roles, and control strategies, consolidating design considerations for practitioners as being presented in [5]. However, as a review, it synthesizes trends but does not deliver a reproducible, station-level optimization workflow that jointly minimizes NPC/LCOE subject to component and SOC constraints. While [6] survey maps EV charging technologies and impacts, including infrastructure trends and control approaches. However, the study lacks quantitative case studies that compare diesel-only vs. PV–diesel with PV–diesel–BESS under realistic 24-hour EV load profiles and tariff/fuel sensitivities. Reference in [7] analyze stand-alone hybrid EV charging with hydrogen storage, showing feasibility in certain climates and resource conditions. But hydrogen pathways are modeled, but practical, near-term PV–BESS–diesel combinations common in developing regions receive less comparative depth in dispatch/risk analysis. Also, [8] provide a systems-approach that review frames EV charging stations holistically (technology, management, and policy layers), offering integrative insights for planners. However, the systems perspective is qualitative; it does not translate into an implementable optimization model that produces concrete size/dispatch recommendations or quantifies LCOE/NPC impacts However, few studies present a reproducible, station-level techno-economic optimization that; jointly sizes PV, diesel, and battery; enforces operational/SOC and reliability constraints tailored to fast-charging; compares diesel-only, PV with diesel, and PV-diesel-BESS on LCOE/NPC under realistic time-series EV demand, tariffs, and fuel prices; and yields actionable dispatch insights. The study therefore, directly targets this gap by formalizing such an optimization and illustrating it with a transparent case that reports dispatch, SOC trajectories, energy shares, and LCOE/NPC outcomes. IJEMT 3. Methodology This tudy focuses on the techno-economic optimization of a hybrid PV–diesel–battery energy system designed for EV charging applications. The objective is to determine the optimal component sizes and dispatch strategy that minimize the overall life-cycle cost of the system while ensuring reliable energy supply to meet the charging demand. The system configuration includes three major components: photovoltaic (PV) arrays, a diesel generator (DG), and a battery energy storage system . Each component contributes to meeting the hourly load demand based on the availability of solar energy, battery state of charge , and generator operational limits. The optimization process simultaneously evaluates the net present cost and levelized cost of energy while satisfying power balance, component, and reliability constraints. 3.1 Objective Function The objective function of the optimization model is to minimize the total Net Present Cost and Levelized Cost of Energy of the hybrid system over its project lifetime, subject to operational and technical constraints. The NPC represents the sum of all capital, replacement, operation, maintenance, and fuel costs throughout the system’s lifespan, discounted to present value as given in Equation (1): ( ) = + + + + = N y y y rep y fuel y M cap u r C C C C NPC t 1 , . , & ) , 1 ; min (1) where NPC is Net Present Cost of the hybrid system, cap C is initial capital investment cost of system components, y M O C , & is annual operation and maintenance cost in year y , y fuel C , is annual fuel cost for the diesel generator in year y , y rep C , is component replacement cost in year y , r is real discount rate (fraction), N is project lifetime , is the vector of decision variables (component sizes and capacities), tu is dispatch decision variables at each time step t (hourly operation). The LCOE represents the average cost per unit of electricity delivered by the hybrid system over its lifetime and serves as a comparative metric between system configurations as given in Equation (2); ( ) = = N y y E N r CRF NPC LCOE 1 , min, (2) where LCOE is levelized cost of energy, ( ) N r CRF , is capital recovery factor, y E is total electrical energy supplied to the load in year y. The objective is therefore to identify component sizes ( ) B DG PV E P P , , = that minimize NPC and LCOE while maintaining high system reliability and minimum unmet load probability. The objective function is subjected to the following technical constraints: i. System power balance constraint: For every time step t, the power supplied from all components must equal the total EV charging demand with losses. The balance equation is given as: Equation (3) t t gris ch t B dis t B t SD use t PV t P P P P P P L exp, , , . , , − + − + + = (3) IJEMT where tL is EV charging load at time t (kW), use t PY P , is PV power used to serve the load (kW), t DG P , is power output from the diesel generator (kW), dis t B P , is battery discharge power at time t (kW), ch t B P , is battery charging power at time t (kW), t grid P , is power imported from the grid (kW), t Pexp, is power exported to the grid (kW) ii. Battery dynamics and operational limits:The state of charge of the battery is updated dynamically based on the charging and discharging activities at each time step as given in Equation (4): B dis dis t B ch t B ch t t E P P SOC SOC , , 1 − + = + (4) Subject to: max min SOC SOC SOC t where t SPC is battery state of charge at time t (%), dis ch , are battery charging and discharging efficiencies (fraction), B E is total energy capacity of the battery (kWh, max min,SOC SOC are minimum and maximum SOC limits iii. Reliability and feasibility constraints To ensure system reliability, the Loss of Power Supply Probability is constrained to approximately zero for fast-charging applications as presented in Equation (5) 0 1 1 = = = T t t T t unserved t L E LPSP (5) where unserved t E represents the unmet energy demand at time t. This ensures that the hybrid system is capable of meeting EV charging requirements under variable solar irradiance and load conditions. The optimization process was carried out using HOMER Grid’s built-in search and simulation engine, which iteratively varies component sizes, evaluates dispatch strategies, and computes performance indicators. The model evaluates multiple scenarios—diesel-only, PV+diesel, and PV+diesel+battery to determine the configuration with the lowest NPC and LCOE. Sensitivity analyses were also conducted on fuel price, battery cost, and solar irradiation to evaluate the robustness of the optimal design. 3.2 Case Study Setup The proposed hybrid PV–Diesel–Battery system was modeled for Agric Village, Ogbomoso, Nigeria, located at latitude 8.1333° N and longitude 4.2667° E. The area experiences a tropical wet and dry (savanna) climate, characterized by two distinct seasons—rainy (April to October) and dry (November to March). Also, the site has high solar irradiance, averaging 5.2 kWh/m2/day, with an annual ambient temperature range of 24–34 °C, conditions that favor photovoltaic generation. The existing grid supply in the area is unreliable, leading to frequent blackouts, which justifies deploying a hybrid system to sustain EV charging operations. A stylized EV charging demand profile was generated based on typical urban mobility patterns IJEMT showing pronounced morning (7–9 a.m.) and evening (5–8 p.m.) peaks corresponding to commuter travel hours. The total daily energy consumption for the modeled station was set at approximately 1,200 kWh/day, distributed across six charging points of mixed power ratings (fast and standard chargers). The system was designed to ensure continuous operation even under cloudy or peak-demand conditions. Three configurations were compared to evaluate techno-economic performance as follows: i. Scenario 1= Diesel-only system: A standalone 250 kW DG supplying all EV charging demand. ii. Scenario 2 =PV + diesel hybrid: A 250 kW PV array integrated with a 200-kW diesel generator. iii. Scenario 3= PV + diesel + battery hybrid: A combination of 250 kW PV array, 200 kW diesel generator, and a 400-kWh lithium-ion battery bank. In all cases, PV generation was prioritized to meet the load, followed by battery discharge, and finally diesel backup for deficit supply periods. The battery bank smoothens the load curve, mitigating generator runtime and improving system efficiency. Cost parameters were adopted based on recent literature and regional market datal PV module cost: $800/kW, Battery storage cost: $300/kWh, Diesel generator cost: $400/kW, Diesel fuel price: $1.35/L, Project lifetime: 25 years, Real discount rate: 8% All simulation parameters, including the solar resource, load data, component costs, efficiency factors, and economic assumptions, were entered into HOMER Grid. The software iteratively optimized system sizes and dispatch strategies to minimize the Net Present Cost and Levelized Cost of Energy while satisfying the Loss of Power Supply Probability (LPSP ≈ 0) requirement. 4. Results and Discussion The simulation and optimization were performed in HOMER Grid for three different configurations: i. Diesel-only system ii. PV + Diesel hybrid system iii. PV + Diesel + Battery hybrid system The optimization process evaluated each configuration in terms of NPC, LCOE, annual energy contribution by source, and battery State of Charge trends. The results are presented and discussed as follows. 4.1. Energy Dispatch and Power Contribution Table 1 presents the annual energy contribution of each power component under the three studied configurations. While a bar graph compares the three scenarios presented in Figure 1, showing that PV integration substantially reduces diesel energy use, and the addition of a battery further smoothens energy delivery, reducing generator operation hours. The PV-diesel-battery configuration achieved the most balanced energy supply, with the PV array meeting 52% of total demand, diesel contributing 22%, and the battery providing 26% through stored energy discharge. This structure ensured continuous charging availability and minimal generator runtime, thereby improving system efficiency. Compared to the diesel-only setup, the hybrid system achieved approximately 72% fuel reduction, highlighting the economic and environmental benefits of renewable integration in rural EV charging systems. Table 1: Annual Energy Contribution and Load Coverage IJEMT Configuration PV Contribution (kWh/year) Diesel Contribution (kWh/year) Battery Contribution (kWh/year) Unmet Load (%) Remarks Diesel-Only 0 438,000 0 0.00 High fuel dependency PV + Diesel 268,000 231,000 0 0.00 Reduced fuel consumption by ~47% PV + Diesel + Battery 285,000 120,000 68,000 (charge/discharge cycles) 0.00 Balanced operation, high reliability Figure 1 (Bar Graph): Energy Contribution by Source 4.2. Battery Performance and State of Charge The battery dynamics for the PV + diesel + battery configuration was evaluated to determine energy storage efficiency and daily SOC variations as given in Table 2. While Figure 4.2 presents the battery SOC profile over 24 hours. The graph shows an increasing SOC from 09:00– 14:00 h (charging from PV) and decreasing from 17:00–22:00 h (discharging to meet evening EV load). The SOC trend indicates that the battery charged optimally during solar hours and discharged during evening peaks, minimizing reliance on the diesel generator. The SOC remained within the safe operational limits (20–95%), which validates both battery capacity adequacy and controller efficiency in scheduling charge/discharge cycles. This operational strategy extended generator life, reduced noise and emissions, and ensured smooth EV charging even under cloudy conditions. The battery autonomy of 2.5 hours also provided short-term backup during transient load surges. IJEMT Table 2: Battery Operation Performance Indicators Parameter Symbol Value Unit Nominal Capacity (E_B) 400 kWh Minimum SOC (SOC_{min}) 20 % Maximum SOC (SOC_{max}) 95 % Round-trip Efficiency (η_{B}) 92 % Daily Charge/Discharge Energy — 68,000 kWh/year Average Battery Autonomy — 2.5 hours Figure 2 (Line Graph): Battery SOC Profile Over 24 Hours 4.3 Economic Evaluation (NPC and LCOE) The economic performance of the three configurations was compared based on Net NPC and LCOE derived from the optimization results as illustrated in Table 3. Figure.3, also shows the comparison of LCOE for three configurations. The bar graph clearly shows a steady reduction in LCOE across configurations: from $0.412/kWh (diesel-only) to $0.227/kWh (PV + Diesel + Battery). The inclusion of PV reduced both NPC and LCOE due to lower fuel consumption and minimal generator operation. Further addition of the battery resulted in an additional 23% LCOE reduction due to optimized dispatch and peak shaving. The PV + Diesel + Battery configuration achieved the lowest NPC ($642,500), offering a 37% total life-cycle cost reduction compared to diesel-only operation. The CO2 emission reduction of over 60% confirms its environmental sustainability. The investment payback period was approximately 6.5 years, making it a financially viable solution for semi-urban charging stations like Agric Village. IJEMT Table 3: Comparative Economic Performance Configuration NPC LCOE (USD/kWh) Annual O&M Cost Fuel Savings (%) CO2 Reduction (%) Diesel-Only 1,025,000 0.412 82,400 — — PV + Diesel 756,000 0.295 54,300 46.2 43.8 PV + Diesel + Battery 642,500 0.227 48,100 67.8 63.5 Figure 3: Comparison of LCOE for Three Configurations 4.4. Sensitivity Analysis Sensitivity analysis was performed to assess the robustness of the optimized PV–diesel–battery system to variations in two critical economic factors diesel fuel price and battery capital cost. These parameters are highly dynamic in the Nigerian energy market, and understanding their impact on system economics is crucial for long-term project sustainability. The results of the analysis were presented in Tables 4 and 5. Table 4 presents the effect of diesel fuel price variation on the hybrid system economics. As the diesel fuel price increased from $1.00 to $1.70 per litre, the NPC rose from $595,000 to $714,200, and LCOE increased by about 19.6%. The rise in LCOE was more pronounced in the diesel-only and PV+diesel configurations, whereas the PV+diesel+battery hybrid exhibited greater economic stability, showing only a moderate increase of about 12.8% at the highest fuel price scenario. This confirms that integrating renewable and storage components enhances resilience to fuel price fluctuations—a critical factor for energy systems in volatile fuel markets like Nigeria. IJEMT Table 4: Effect of Diesel Fuel Price Variation on System Economics Diesel Price (USD/L) NPC LCOE (USD/kWh) Fuel Savings Compared to Diesel-Only (%) Change in LCOE (%) CO2 Emission Reduction (%) 1.00 595,000 0.214 70.5 –5.7 61.0 1.10 618,000 0.220 69.1 –3.1 60.4 1.20 642,500 0.227 67.8 0.0 59.8 1.35 (Base Case) 667,800 0.236 66.2 +3.9 59.0 1.50 689,900 0.245 65.0 +7.9 58.4 1.70 714,200 0.256 63.5 +12.8 57.1 Table 5 also present the effect of battery cost variation on designed hybrid system economics. A progressive decrease in battery cost from $400/kWh to $200/kWh produced a 10.7% reduction in LCOE and shortened the payback period by approximately 1.6 years. This clearly demonstrates that declining global battery prices significantly improve the economic competitiveness of hybrid EV charging systems. Conversely, a 30% rise in battery cost increases the LCOE only marginally (+5.3%), showing that while battery cost influences system economics, its effect is less severe than that of fuel price fluctuations. The results demonstrate that a PV + Diesel + Battery hybrid system is the most technically reliable, economically feasible, and environmentally sustainable configuration for powering EV charging infrastructure in Agric Village, Ogbomoso. It successfully balances renewable intermittency, reduces fuel dependency, and provides a stable power supply in a region with erratic grid conditions. These findings reinforce global trends emphasizing hybrid renewable microgrids as the future of rural EV charging and energy transition in sub-Saharan Africa. Table 5: Effect of Battery Cost Variation on System Economics Battery Cost (USD/kWh) NPC LCOE (USD/kWh) Fuel Savings (%) Payback Period Change in LCOE (%) 400 695,500 0.239 65.4 7.3 +5.3 350 670,800 0.231 66.8 6.9 +1.7 300 (Base Case) 642,500 0.227 67.8 6.5 0.0 250 625,000 0.219 68.6 6.1 –3.5 200 598,000 0.214 70.1 5.7 –5.7 5. Conclusion This study developed and analyzed a techno-economic optimization model for a hybrid PV– Diesel–Battery energy system designed to supply reliable power for electric vehicle (EV) charging at Agric Village, Ogbomoso, Nigeria. Using HOMER Grid, three configurations— Diesel-only, PV + diesel, and PV + diesel + battery were simulated to evaluate performance based on NPC, LCOE, and system reliability. Results revealed that the PV + diesel + battery configuration achieved the lowest LCOE (0.227 USD/kWh) and NPC (642,500 USD), with IJEMT significant reductions in fuel consumption (≈ 68 %) and carbon emissions (≈ 63 %) compared to the diesel-only setup. Battery integration enhanced operational flexibility by storing midday solar energy and supplying evening peak demand, while sensitivity analyses demonstrated that the system remains economically resilient under fluctuations in fuel and storage costs. Overall, the optimized hybrid system offers a sustainable, cost-effective, and dependable solution for EV charging infrastructure in regions with unstable grid conditions, supporting Nigeria’s transition toward cleaner transportation and renewable energy development. IJEMT 6. References 1. Güven, A. F., Özgür, M., & Alkan, M. (2025). Sustainable hybrid systems for electric vehicle charging stations. Scientific Reports, 15(1), 87985. 2. Okafor, C. E., Nwodo, C. O., & Ike, V. A. (2025). Techno-economic analysis of battery storage technologies in grid-connected PV systems with EVs. Sustainable Energy Technologies and Assessments, 67, 103228. 3. Karmakar, A., & Banerjee, S. (2025). Techno-economic analysis and optimized PV- powered EV charging facilities under various climate conditions in India. Discover Sustainability, 6(1), 55–68. 4. Mousa, M. M., Ali, M., Abdel-Rahman, M., & Farag, H. E. (2025). Optimizing grid-tied hybrid renewable systems for EV charging. Energy Reports, 11(2), 712–726. 5. Ramkumar, G., Venkatesh, K., & Prakash, P. (2025). A review of renewable energy systems integration for EV charging facilities. Energy Reports, 11(4), 1259–1274. 6. Sarda, J., & Kalband, K. (2024). A review of EV charging technology, impact, and trends. Sustainable Energy Technologies and Assessments, 65, 103071. 7. Soomro, A. M., Shah, S. A., & Memon, A. M. (2024). Techno-economic analysis of stand-alone hybrid PV-hydrogen EV charging station. Sustainable Energy Technologies and Assessments, 64, 103042. 8. Tejeida-Padilla, R., Hernández-Lemus, E., & Vargas-Bello-Pérez, E. (2024). A comprehensive review of electric charging stations from a systems approach. World Electric Vehicle Journal, 15(12), 571. 9. International Energy Agency . (2025). Global EV Outlook 2025: Electric vehicle charging. Paris: IEA Publications. 10. U.S. Department of Energy . (2024). EVs@Scale FY 2023 Year in Review (Vehicle Technologies Office report). Washington, DC: DOE. https://energy.gov/vehicles 11. HOMER Energy. (n.d.). HOMER Pro User Manual v3.7. https://homerenergy.com/pdf/HOMERHelpManual.pdf (Accessed August 17, 2025). 12. HOMER Energy. (n.d.). Optimization in HOMER. https://www.homerenergy.com/products/pro/docs/3.15/optimization.html (Accessed August 17, 2025).