RNN-Based Control Strategy for Grid Connected Hybrid Systems Integrating Renewable Energy Sources

This paper introduces an innovative RNN-based control approach for grid connected systems that incorporate Renewable Energy Sources (RES). The approach utilizes Recurrent Neural Networks (RNNs) to optimize power management and enhance the stability of systems featuring solar panels with an Improved LUO Converter for enhancing the voltage along with fuel cells utilizing a boost converter. By harnessing RNN based Maximum Power Point Tracking (MPPT), system accurately predicts and adapt to fluctuations in energy output. To verify the robustness of the proposed approach, key performance measures such as response time, energy loss, and grid stability are examined. The goal of this project is to develop an advanced control framework that tackles the issues of energy reliability and variability while successfully integrating a variety of renewable sources. After evaluating the suggested design, simulation results on the MATLAB platform demonstrate a 91% MPP tracking efficiency and a decreased THD of $\mathbf{0. 8 9 \%}$.