Optimal dispatch strategy analysis for PV/diesel/battery hybrid power systems in remote areas using MATLAB simulation

This study presents a detailed simulation-based assessment of hybrid PV/diesel battery hybrid power systems designed for remote off-grid applications in Thailand’s national parks. A MATLAB/Simulink model was developed to evaluate systems of varying scales, combining 15–50 kWp photovoltaic arrays, 30–60 kW diesel generators, and 250–800 kWh battery banks. The model incorporated measured irradiance and temperature data, dynamically calculating PV performance and tracking the battery state of charge (SOC) between 20% and 100%. Three dispatch strategies were applied: Cycle Charging (CC), Load Following (LF) and Predictive Dispatch (PS) to assess their impact on energy balance, fuel use and PV curtailment. The CC dispatch strategy runs the diesel generator at full load to recharge the batteries rapidly, enhancing generator efficiency but increasing battery cycling and PV curtailment. The LF dispatch strategy operates the generator only when PV power is insufficient to meet the demand, which reduces generator runtime but often results in lower fuel efficiency. The PS strategy utilizes short-term forecasts of solar irradiance and load to schedule generator operation and battery charging, thereby minimizing start–stop cycles and maximizing PV utilization.

The simulation results revealed distinct operational patterns, with clear differences in fuel consumption, SOC behavior, and clipping losses for each dispatch strategy. Figure 1 shows MATLAB simulation outputs comparing Cycle Charging, Load Following, and Predictive Dispatch strategies, illustrating how each approach impacts generator operation, battery cycling, and PV utilization in the off-grid system. From Figure 12, it is evident that the most suitable dispatch strategy for the current off-grid system is the ideal PS (Predictive Strategy). The PS strategy continuously forecasts system operation, resulting in annual diesel consumption of 6,984 liters and energy production of 382 kWh/year. Under the LF (Load Following) strategy, diesel consumption is higher because the generator is frequently operated at partial load, while the CC (Cycle Charging) strategy produces even greater fuel use since it charges the battery fully every time it runs (including energy lost from charging and inverter conversion). In the PS strategy, the generator starts less frequently but operates longer during each run, reducing the number of starts and stops, and optimizing fuel use. According to the study, PS led to 370 annual generator starts with a total annual fuel consumption of 7,544 liters/year, whereas CC resulted in annual fuel use of 3,469 kWh/year and LF caused the highest fuel consumption at 7,852 liters/year. These findings provide a technical roadmap for scaling hybrid systems in Thailand’s national parks and expanding them to other remote conservation areas nationwide. This research was financially supported by the National Research Council of Thailand (NRCT).