Many remote communities rely on diesel generators as their primary power source, which is expensive and harmful to the environment. Renewable energy systems, based on photovoltaics and wind turbines, present a more sustainable and potentially cost-effective option for remote communities with abundant sun and wind. Designing and implementing community-owned and operated renewable power generation alternatives for critical infrastructure such as hospitals, water sanitation, and schools is one approach towards community autonomy and resiliency. However, configuring a cost-effective and reliable renewable power system is challenging due to the many design choices to be made, the large variations in the renewable power sources, and the location specific renewable power source availability. This paper presents an optimization-based approach to aid the configuration of a solar photovoltaic (PV), wind turbine generator and lead-acid battery storage hybrid power system. The approach, implemented in MATLAB, uses a detailed time-series system model to analyze system Loss of Load Probability (LOLP) and a lifetime system cost model to analyze system cost. These models are coupled to a genetic algorithm to perform a multi-objective optimization of system reliability and cost.
The method was applied to two case studies to demonstrate the approach: a windy location (Gibraltar, UK), and a predominantly sunny location (Riyadh, Saudi Arabia). Hourly solar and wind resource data was extracted for these locations from the National Oceanic and Atmospheric Administration for five-year data sets. The village load requirements were statistically generated from a mean daily load for the community estimated based on the population and basic electricity needs. The case studies demonstrate that the mix and size of technologies is dependent on local climatic conditions. In addition, the results show the tradeoff between system reliability and cost, allowing designers to make important decisions for the remote communities.