A convolutional neural network based energy management system for photovoltaic/battery systems in microgrid using enhanced coati optimization approach

The study proposes a centralized novel approach for energy management system (EMS) based on Convolutional Neural Networks (CNNs) designed for microgrids that incorporate hybrid PV and BESS DGs. Additionally, a CNN-EMS has been created that enables near-real-time SoC balancing by using NNs to learn the nonlinear mapping between the cause and the control action. The trained NN-based central controller does not need to do intricate calculations to forecast nearly optimal control actions in virtually real-time. The CNN weights are optimized using a novel Enhanced Coati Optimization (E-COA) approach. This method improves the accuracy and efficiency of the CNN-based EMS and the accompanying error measurements are predicted. The experimental assessment and analysis will be done, and the performance of the provided technique will be simulated using MATLAB. The results show that proposed method outperforms existing techniques in terms of error reduction. Specifically, when comparing learning percentages across various error indices, proposed method achieved the lowest Mean Absolute Error (MAE) values: 8.23 % at 60 %, 3.08 % at 70 %, and 2.81 % at 80 %, all of which are lower than those of existing methods. These findings demonstrate that the proposed system significantly improves forecasting accuracy and prediction effectiveness.