Load forecast method of electric vehicle charging station using SVR based on GA-PSO

3rd International Conference on Advances in Energy, Environment and Chemical Engineering

能源学;生态环境科学;化学工业

Lu, Kuan^1 ; Sun, Wenxue^2 ; Ma, Changhui^1 ; Yang, Shenquan^3 ; Zhu, Zijian^3 ; Zhao, Pengfei^3 ; Zhao, Xin^3 ; Xu, Nan^3

State Grid Shandong Electric Power Research Institute, Jinan

250002, China^1

State Grid Zhangqiu Power Supply Company, Jinan

250200, China^2

EconomicandTechnology Institute, State Grid Shandong Electric Power Company, Jinan

250021, China^3

关键词: Charging station;    Electric vehicle charging;    Ev charging stations;    Forecasting accuracy;    Fuzzy C means clustering;    Global parameters;    Local searching;    Support vector regression (SVR);   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/69/1/012196/pdf
DOI  :  10.1088/1755-1315/69/1/012196

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

This paper presents a Support Vector Regression (SVR) method for electric vehicle (EV) charging station load forecast based on genetic algorithm (GA) and particle swarm optimization (PSO). Fuzzy C-Means (FCM) clustering is used to establish similar day samples. GA is used for global parameter searching and PSO is used for a more accurately local searching. Load forecast is then regressed using SVR. The practical load data of an EV charging station were taken to illustrate the proposed method. The result indicates an obvious improvement in the forecasting accuracy compared with SVRs based on PSO and GA exclusively.