Solar and Wind Energy Forecasting for Green and Intelligent Migration of Traditional Energy Sources

Fossil-fuel-based power generation leads to higher energy costs and environmental impacts. Solar wind are abundant important renewable sources (RES) that make the largest contribution replacing fossil-fuel-based consumption. However, uncertain solar radiation highly fluctuating weather parameters of require an accurate reliable forecasting mechanism for effective efficient load management, cost reduction, green environment, grid stability. From existing literature, artificial neural networks (ANN) a better means prediction, but ANN-based techniques lose prediction accuracy due high uncertainty input data random determination initial weights among different layers ANN. Therefore, objective this study is develop harmony search algorithm (HSA)-optimized ANN model energy. In study, we combined with HSA provided feedback its adjustment HSA, instead Then, optimized were assigned edges weights, completes training Extensive simulations carried out our proposed HSA-optimized irradiation forecast achieved values MSE = 0.04754, MAE 0.18546, MAPE 0.32430%, RMSE 0.21805, whereas speed 0.30944, 0.47172, 0.12896%, 0.55627. Simulation results prove supremacy models compared state-of-the-art techniques.