Gray prediction and support vector machine-based classification type electric vehicle demand temporal-spatial distribution dynamic prediction method

Abstract

The invention is applied in the field of electric power systems, and in particular relates to a method for dynamic forecasting of demand for classified electric vehicles based on gray forecasting and support vector machines. Including: firstly, use the high-precision improved gray model to predict the number of different types of vehicles; then, based on the proportion of different types of electric vehicles and the nonlinear characteristics of the influencing factors, use the support vector machine regression method to obtain the classification by using the prediction samples Electric vehicles replace the proportional forecast results, and use the iterative method to continuously revise the forecast results; finally, match the first two forecast results according to vehicle types, establish a demand growth forecast model for electric vehicles by type, and combine the research on user travel patterns to determine the demand for electric vehicles by type. Accurate dynamic spatiotemporal forecasting is achieved. Therefore, the present invention has the following advantages: fully considering the characteristics of insufficient historical data and the influence of different factors on the development of electric vehicles, combined with the research on user travel rules, to achieve more accurate dynamic prediction.

Description

technical field [0001] The invention is applied in the field of electric power system, and is used for accurately and reasonably predicting the quantity demand of classified electric vehicles, so as to facilitate the accurate analysis of the charging load of electric vehicles and the reasonable planning and construction of charging stations, and specifically relates to a method based on gray prediction and A Categorical Approach to Electric Vehicle Demand Dynamics Forecasting with Support Vector Machines. Background technique [0002] The energy industry is the basic industry of the national economy, a necessary prerequisite for ensuring national strategic security, and an important guarantee for sustainable economic development. However, with the continuous expansion of my country's economic scale, my country's demand for traditional energy such as oil has not only increased, but also more and more carbon content has been discharged into the atmosphere, which has increased ...

AI Technical Summary

Problems solved by technology

[0005] At present, domestic scholars have conducted a lot of research on the prediction of electric vehicle ownership, but the consideration factors are relatively single, and they rarely conduct classification research. They mainly use a single method such as linear regression to estimate, which is accurate in predicting the classification of electric vehicle ownership. There is a certain influence on the characteristics of electric vehicles, and the research on the dynamic prediction of the temporal and spatial distribution of electric vehicle demand by type is still blank.

Images