Artificial intelligence (AI) in renewable energy technologies plays a crucial part due to its modeling and performance forecasting. Therefore, a novel AI-based evolving generative adversarial Fuzzy network (EGAFN) has been presented in this paper as a forecasting tool for the efficiency analysis of renewable solar energy for four distinct regions. The power ratings from environmental parameters and solar panels were monitored and recorded for a year. The data pre-processing is primarily applied to improve the system's function using a data filter. Furthermore, the data's energy estimation accuracy is enhanced using feature extraction and selection by a multi-objective lionized manta-ray foraging optimizer (MLMRFO). Finally, the hyperparameters of the EGAFN method are optimized by multi-objective optimization. The proposed technique aims to enhance the energy efficiency of PV systems for solar power production forecasting using an optimized multi-objective algorithm. The findings show that the suggested technique's prediction performance is better than earlier methods. Thus, the proposed methodology can assist in increasing energy efficiency and making better use of renewable energy sources.