Artificial Intelligence-Based Prediction of Renewable Energy Sources Using Correlation Testing

The purpose of renewable energy to generate power and change the composition of the energy mix is becoming more popular in today's world. The competition of solar power plants in the energy market and decreasing reliance on fossil fuels for socio andeconomic growth are both facilitated by solar energy forecasting, which is a critical component. To overcome this issue, we presented an Adaptive sea lion optimized genetic adversarial network (ASLO-GAN) method. The purpose of the ASLO-GAN method is to predict the renewable energy sources. The dataset of renewable energy sources (RESs) is collected from kaggleand then the collected dataset is preprocessed using decimal scale normalization. When extracting the data, the spearman rank-order correlation (SROC) method is utilized so undesirable data can be omitted from the process. The simulation results include RMSE, MAPE, R-square, CV(RMSE) and NMBE that are evaluated. The main key findings of NMBE were compared to other traditional approaches in order to show the validity and repeatability of the inquiry.