Novel Methodology to Measure the Reliability of Public DC Fast Charging Stations

Abstract

A network of reliable corridor charging stations is essential to building driver confidence in long-distance battery electric vehicle trips. Here, we propose a detailed methodology to measure station reliability based on charging infrastructure data. By assigning charging events to unique charging visits, our methodology can capture a holistic overview of the driver’s charging experience. We use real world charging data collected between 2019 and 2022 from 54 Direct Current Fast Chargers (DCFCs) in 36 corridor charging stations across California to demonstrate that our overarching reliability framework is close to the experience of users. Our analysis of two different charging networks shows that users of these networks have an average chance of 83% and 77% generally successful outcomes, respectively, after one or more tries at a charging station location. We also find significant variation in station performance within the same network (i.e., 79%–87% and 13%–95%, respectively). Furthermore, we observe that at least 3% of users are facing unexpected charging interruptions. In addition, we demonstrate a practical application of our framework for deep diagnostics of the charging eco-system using error codes to identify common issues such as vehicle/charger communication issues, safety issues, payment issues, and cable/connector issues. We compare how error codes alone are not a good proxy to diagnose charging failures. As more data from DCFCs becomes available, our methodology can become a mainstream tool for evaluating station reliability.