Spiking Deep Residual Network Optimized Using Pied Kingfisher Optimizer for Renewable Energy Forecasting in Microgrids

Renewable energy forecasting in Microgrids (MGs) enables efficient power management by predicting energy generation from sources like wind and solar, helping to balance supply and demand. However, wind power forecasting faces challenges due to the intermittent and highly variable nature of wind speed, leading to potential errors in prediction. Additionally, uncertainties in meteorological conditions, complex terrain effects, and limited high-resolution historical data can impact forecasting reliability, affecting overall MG performance. To overcome these drawbacks, this manuscript proposes a renewable energy forecasting in MG for predict short-term wind power using SDRN-PKO approach. The data are collected from Woolnorth Wind Site Data in Australia. Afterward, the data are fed to pre-processing. In pre-processing segment removes the missing values and normalization in the data utilizing Maximum Correntropy Quaternion Kalman Filter (MCQKF). The pre-processed output was fed to Spiking Deep Residual Network (SDRN) for predicting short-term wind power of MG. The Pied Kingfisher Optimizer (PKO) is used to optimize the weight parameter of SDRN. The proposed SDRN-PKO is utilized within the MATLAB platform. The proposed SDRN-PKO technique is compared with the existing techniques such as Recurrent Neural Network-Gated Recurrent Unit (RNN-GRU), Long Short-Term Memory-Gated Recurrent Unit (LSTM-GRU), and Deep Reinforcement Learning-Teaching Learning based Optimization (DRL-TLBO) respectively. Performance metrics including Root Mean Squared Error (RMSE), Mean Squared Error (MSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE) is examined in order to determine the proposed method. The SDRN-PKO method achieves a MAPE of 15, MAE of 12, MSE of 12, and a RMSE of 10, demonstrating its superior performance in predicting short-term wind power. The SDRN-PKO method's lower error rates, coupled with its robust performance, make it a reliable and efficient solution for wind power forecasting in MGs.