Predicting the Availability of Supply Equipment at EV Charging Stations

Sun，Chenxi1; Sun，Haochen2; Jia，Youwei3; Tang，Xiaoying1,4

10.1109/EI256261.2022.10116868

2022

979-8-3503-4716-6

EI2 2022 - 6th IEEE Conference on Energy Internet and Energy System Integration

1628-1633

11-13 Nov. 2022

Chengdu, China

Understanding and characterizing Electric Vehicle (EV) users' behavior at EV charging stations is beneficial for improving the operation efficiency of EV charging stations through smart scheduling. Recently, the fast development of EV charging infrastructure worldwide and the release of collected real EV charging dataset make it possible to learn from actual observations in practice. In this paper, we investigate the possibility of predicting the usage of electric vehicle supply equipment at EV charging stations using historical user data and weather data. We formulate the prediction problem as a time-series prediction problem. We evaluate and compare the performance of two popular neural network models against traditional methods for predicting the usage of Electric Vehicle Supply Equipment (EVSE) at charging stations using historical data. Experiments show that for workplace charging sites, these neural network models outperform baseline methods and are capable to predict the vacancies of EVSE with high accuracy.

electric vehicle (EV) long short term memory network predictive model time series

通讯

英语

2-s2.0-85160208352

Scopus

1.Shenzhen Institute of Artificial Intelligence and Robotics for Society,Shenzhen,China
2.The Chinese University of Hong Kong (Shenzhen),School of Science and Engineering,Shenzhen,China
3.The Chinese University of Hong Kong (Shenzhen),School of Data Science,Shenzhen,China
4.Southern University of Science and Technology,Department of Electrical and Electronic Engineering,Shenzhen,China

Sun，Chenxi,Sun，Haochen,Jia，Youwei,et al. Predicting the Availability of Supply Equipment at EV Charging Stations[C],2022:1628-1633.