Analysis of the Influence of Meteorological Factors on Electric Vehicle Charging Load and Mixed Regression Forecasting Model

Abstract

AbstractElectric vehicle charging load forecasting is the basis for carrying out research on electric vehicle charging infrastructure planning and construction, power system optimization and scheduling, etc. Since the electric vehicle charging behavior is random in terms of time and space, a lot of complicated factors can influence the charging load forecasting, and different charging load forecasting models and results will be obtained if it is considered from different perspectives. In this paper, the author first researches the influence of meteorological factors on electric vehicle charging load with the correlation analysis method, and particularly distinguishes the difference between the influence of meteorological factors on conventional electric load and on electric vehicle charging load. The analysis results reveal that the response of electric vehicle charging load to meteorological factors is relatively lagging behind, and thus the author proposes a mixed regression forecasting model based on meteorological factors. In this model, indicators of multiple meteorological factors are considered and the idea of similar date method is combined. The example analysis shows the forecast accuracy of this model is obviously higher than that of the single factor forecasting model. Finally, this paper further points out the possible research direction of electric vehicle charging load based on meteorological factors.KeywordsElectric vehicleLoad forecastingCorrelation analysisMixed regression model