A PSO-ANN Hybrid Model for Electric Vehicle Charging Demand Forecasting for EV Supply Chain

A PSO-ANN Hybrid Model for Electric Vehicle Charging Demand Forecasting for EV Supply Chain

Pavan Vishwakarma1, Prof. Manish Soni2

Department of Mechanical Engineering,1,2
MIT, Ujjain, India.1,2

ABSTRACT: With the high frequency roll out of electric vehicles, simultaneous charging may pose a challenge in the near future, especially in peak hours. Hence it is mandatory to analyze previous data pertaining to charging demand for electric vehicles to meet the charging and infrastructure demands in the future. This will allow both the utility providers as well as the consumers to plan for peak charging times and associated congestion at EV charging stations. Several researchers have explored data driven machine learning models to predict EV charging demands as machine learning models can analyze complex patterns and relationships within the data to generate accurate predictions. This paper presents a Particle Swarm Optimization (PSO)-Neural network model for EV charging demand prediction. Contrary to conventional neural network models, the PSO is used to adaptively update the network weights. The results clearly indicate that the proposed approach outperforms existing baseline approaches in terms of prediction MAPE.

Keywords: Electric Vehicles, EV Charging Demand, Demand Forecasting, Particle Swarm Optimization (PSO), Deep Neural Networks.