Data-Driven Thermal Modelling for Anomaly Detection in Electric Vehicle Charging Stations

TL;DR

This paper introduces a data-driven, neural network-based approach for detecting anomalies in EV charging stations by modeling thermal behavior to enable condition monitoring and predictive maintenance.

Contribution

It presents a novel human-interpretable thermal modeling method combined with machine learning for anomaly detection in EV charging infrastructure.

Findings

Effective neural network models identified anomalies accurately.

Thermal modeling enabled predictive maintenance insights.

Improved reliability of EV charging stations.

Abstract

The rapid growth of the electric vehicle (EV) sector is giving rise to many infrastructural challenges. One such challenge is its requirement for the widespread development of EV charging stations which must be able to provide large amounts of power in an on-demand basis. This can cause large stresses on the electrical and electronic components of the charging infrastructure - negatively affecting its reliability as well as leading to increased maintenance and operation costs. This paper proposes a human-interpretable data-driven method for anomaly detection in EV charging stations, aiming to provide information for the condition monitoring and predictive maintenance of power converters within such a station. To this end, a model of a high-efficiency EV charging station is used to simulate the thermal behaviour of EV charger power converter modules, creating a data set for the training of neural network models. These machine learning models are then employed for the identification of anomalous performance.