دانلود رایگان مقاله مدل پیش بینی تقاضای شارژ خودرو الکتریکی بر اساس فناوری داده های بزرگ

Abstract

This paper presents a forecasting model to estimate electric vehicle charging demand based on big data technologies. Most previous studies have not considered real-world traffic distribution data and weather conditions in predicting the electric vehicle charging demand. In this paper, the historical traffic data and weather data of South Korea were used to formulate the forecasting model. The forecasting processes include a cluster analysis to classify traffic patterns, a relational analysis to identify influential factors, and a decision tree to establish classification criteria. The considered variables in this study were the charging starting time determined by the real-world traffic patterns and the initial state-of-charge of a battery. Example case studies for electric vehicle charging demand during weekdays and weekends in summer and winter were presented to show the different charging load profiles of electric vehicles in the residential and commercial sites. The presented forecasting model may allow power system engineers to anticipate electric vehicle charging demand based on historical traffic data and weather data. Therefore, the proposed electric vehicle charging demand model can be the foundation for the research on the impact of charging electric vehicles on the power system.

5. Conclusions

This paper proposed an EV charging demand forecasting model with big data technologies. The historical real-world traffic distribution and weather condition data processed by big data technologies were considered in the EV charging demand forecasting model. These big data handling processes include a cluster analysis to classify traffic patterns on each cluster, a relational analysis to identify influential factors affecting the traffic patterns, and a decision tree to establish classification criteria. The example case studies to anticipate EV charging demand in the residential and commercial sites on weekdays and weekends in winter and summer seasons were presented in Section 4 using the proposed EV charging demand forecasting model. High EV charging demand in the residential sites was observed during the night on weekends because all cars were charged at home on weekends as discussed in Section 3.6. In the commercial sites, high charging demand was observed during the non-operational hours due to its lesser charging starting time interval in this period than that in the other charging periods. The proposed EV charging demand forecasting model in this paper may help on the research on the impact of EV charging demand on the power system. In addition, the proposed EV charging demand forecasting model may allow utility operators to plan the operation and generation profiles in the future power systems by predicting the EV charging demand in the residential and commercial sites. The presented EV charging demand model can also contribute to deciding investment and operation plans for adaptive EV charging infrastructures depending on EV charging demand.