Journal of Engineering Research and Reports

Hybrid energy systems that integrate solar power with other renewable or conventional sources provide a robust solution to intermittency and reliability challenges in Nigeria. This study evaluates the performance of such systems using 1,000 hours of operational data and key performance indicators—supply, demand, losses, and battery state of charge (SoC). Statistical and machine-learning methods, including regression analysis and forecasting models (ARIMA, Prophet, LSTM), were applied. Results show that solar and wind supplied the majority of energy, with the grid primarily serving as backup, while excessive grid reliance increased losses. Forecasting indicated that LSTM outperformed ARIMA and Prophet (R² = 0.91, RMSE = 3.10, MAPE = 5.1%). Optimization further reduced losses and improved battery performance. The study demonstrates that solar-based hybrid systems enhance reliability, efficiency, and cost-effectiveness, providing actionable insights for energy planners and policymakers to promote sustainable, data-driven electrification in Nigeria.