Research on charging load prediction technology of electric vehicles based on multilayer perceptron

Abstract

In recent years, the electric vehicle market has been expanding rapidly. By the end of 2021, the number of new energy vehicles in the country will reach 7.84 million, accounting for 2.60% of the total number of vehicles, an increase of 59.25% over 2020. A small number of electric vehicles will not have an impact on the power grid. But in the future, with the electrification of all passenger cars, large-scale electric vehicles will emerge in the future. The access to large-scale electric vehicles and charging piles to the power grid will bring new challenges to the normal operation and control of the power grid. Therefore, charging load forecasting is essential. This paper proposes a load forecasting model based on a multilayer perceptron. It can predict the charging load of electric vehicles in a specific area and provide a reference for urban infrastructure planning and construction, optimal power flow of power systems, and economic dispatch of power grids.