Efficient EV Fleet Management using Automated Thermal Stress Computation for Transformer in Electrical Distribution System

This paper focuses on the challenge of accommodating the growing share of distributed energy resources (DERs) and electric vehicles (EVs) in electrical distribution systems (EDS). To address this challenge, advanced data-driven fleet analysis is required to integrate EVs and DERs within existing infrastructure while ensuring efficient asset integrity and lifespan. We use a standard IEEE-13 bus system to simulate time series data, which is used to train Long Short-Term Memory (LSTM) neural networks. The proposed approach is used to predict the stress on assets for the upcoming years, and the results are plotted for each day to analyze the lifespan degradation of assets. Through this EV fleet analysis, EDS operators could better understand the impact of EVs and DERs on their systems, which helps them to plan effective asset management strategies. Overall, we emphasize the need for advanced data-driven methods to optimize the integration of EVs and DERs in EDS to ensure reliable and efficient energy delivery to consumers without requiring infrastructure upgrades or compromising the lifespan of assets.