Mid-Term Load Forecasting for Electric Vehicle Charging Stations Based on LASSO-BPNN

• Published in: 2024 IEEE 2nd International Conference on Control, Electronics and Computer Technology (ICCECT) pp. 630 - 635
• Main Authors: Ma, Yifan, Yu, Zhiyun, He, Zhizhuo, Zhao, Junzhe
• Format: Conference Proceeding
• Language: English
• Published: IEEE 26.04.2024
• Subjects: Analytical models; BP neural network; Data models; Electric vehicle charging; EV charging load forecasting; LASSO; Load forecasting; multi-feature; Power system stability; Predictive models; Stability analysis
• DOI: 10.1109/ICCECT60629.2024.10546219

The rapid development of electric vehicles (EVs) posed a significant challenge to the stability of the power grid. Forecasting mid-term load for EV charging stations can assist grid management and operations. This paper proposes a new method based on LASSO-BPNN to forecast mid-term load. Firstly, a multi-feature indicators system was established, including environmental, economic, and societal factors. Secondly, this paper applied Least Absolute Shrinkage and Selection Operator (LASSO) for feature selection to prevent overfitting. Thirdly, considering the indicators selected from LASSO, Back-Propagation Neural Network (BPNN) was used to predict the mid-term load for EV charging stations. Finally, this paper relied on the EV stations charging load data and multi-feature indicators data in Qingpu District, Shanghai, trained the overall model and acquired the forecast results. The results showed that the LASSO-BPNN model exhibited high forecast accuracy through error comparison analysis among different models.