Improving Microgrid Reliability and Efficiency Through Energy Storage Systems with a Pyramidal Dilation Attention Convolutional Neural Network for Renewable Energy Integration

Microgrid (MG) design and deployment are heavily reliant on reliability. Integrating renewable energy sources (RESs) into power grids poses considerable challenges for maintaining grid reliability due to their intermittent nature. This paper proposes a novel approach to enhance grid stability by incorporating Energy Storage Systems (ESS) coupled with Pyramidal Dilation Attention Convolutional Neural Network (PDACNN). The goal of the proposed method is to improve efficiency and lower electricity costs. The proposed PDACNN optimizes the control of Energy Storage Systems by providing more trustworthy forecasts of energy demand as well as supply. By, then the proposed method is implemented in MATLAB and evaluated their performance with various existing methods. The PDACNN method surpasses existing techniques in both efficiency and cost-effectiveness. It achieves a high efficiency of 95.7<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup> and reduces total electricity costs to $650/kWh, outperforming the Artificial Neural Network (ANN), and Recurrent NN (RNN) and Convolutional NN (CNN) methods.