Electric vehicle charging station fault detection: a machine learning approach

The number of electric vehicles on the roads is increasing, and the number of fast DC charging stations is following this trend. However, the advantages of DC electric vehicle charging stations (ECVS) in terms of efficiency and power output compared to their AC counterparts are accompanied with protection problems. To address these problems, a recurrent neural network-based fault detection method that can detect both solid and arc faults in EVCS is presented.