In recent years, the fast expansion of renewable energy generation in the power grid, particularly from wind and solar energy resources, has made these generators a major source of unpredictability. Generation and load balancing are critical in the economic scheduling of manufacturing units and energy market activities. Solar and wind energy projections attract the scientific community and numerous research articles are provided. For solar and wind power short-term forecasting (STF), this paper proposes a resilient back propagation neural network (RBPN) model. Because solar irradiation and wind speed are not linear and unexpected, STF is difficult to complete under changing weather circumstances. However, a RBPN is presented and is appropriate for STF modeling. It also improves power quality in various situations, including voltage imbalance correction, active and reactive power control, and voltage regulation. Simulations performed with MATLAB Simulink software are used to validate the performance of the proposed forecasting system. The suggested method also includes a sensitivity analysis of numerous input variables for the optimal model selection and model performance comparison with multiple linear regression and persistence models. The root mean squared error (RMSE) and mean absolute error (MAE) of proposed wind forecasting are 4.60 and 4.30 respectively. The RMSE and MAE of proposed solar forecasting are 3.02 and 3.10 respectively.