Hybrid Artificial Intelligence Optimization for Solar-Wind Hybrid Energy System Grids

Abstract

One potential route to creating reliable and ecological power systems are to integrate energy from renewable resources into intelligent networks. However, the optimal operation of mixed green power sources remains a crucial field requiring in-depth investigation. This work proposes a comprehensive technique that blends AI algorithmic approaches with metaheuristic optimization algorithms to forecast and control sources of clean energy in intelligent grid settings. The suggested HSTM-RL-PPO paradigm outperforms existing models with enhanced [37] precision, recall, and accuracy scores of 0.93, 0.94, and 0.93, correspondingly, in terms of correctly predicting trends in energy consumption. With a success rate of 0.92 on numerous parameters, the TRPO-RL-SA technique is a useful tool for evaluating load balancing. The CNN-PSO technique is particularly actual at forecasting the production of clean energy, with average squared error (MSE), average absolute error (MAE), root average square error (RMSE), R-squared rating, average absolute percentage error (MAPE), and average squared error (RMSE) [37] corresponding to 345.12, 15.07, 0.78, 18.57, and 7.83, respectively. The outcomes of our study contribute to the advancement of hybrid renewable energy sources within intelligent grid systems, leading to enhanced reliability, efficiency, and cost-effectiveness in energy generation and transmission. Additionally, the proposed method appears applicable in remote and off-grid locations. To sum up, our research establishes a valuable framework for enhancing Promoting renewable energy generation, this study acts as a catalyst, encouraging additional exploration into the energy managing domain