Electric vehicle charging demand prediction method and device

Abstract

The embodiment of the invention provides an electric vehicle charging demand prediction method and device which are applied to an electric vehicle charging demand prediction server. The method comprises the following steps: constructing a test data set; training a load prediction model through the test data set; and inputting the test data set into the trained load prediction model to obtain a prediction result. According to the deep neural network electric vehicle charging demand prediction method based on improved probability sparse self-attention and comparative learning, the problems that the long-term demand prediction error is large and extra information correction is needed can be solved; the method can meet the deployment requirements of mobile terminal use, edge equipment operation, remote area local area network, network security limitation and other scenes, especially can be deployed in a charging station energy storage system or even in a charging pile to independently operate, provides a more accurate prediction result and a more flexible deployment mode, and enables an existing charging station to have an intelligent upgrading space.

Description

technical field [0001] The present application relates to the technical field of electric vehicles, in particular to a method and device for predicting charging demand of electric vehicles. Background technique [0002] In recent years, the electric vehicle industry has developed rapidly, and the large-scale electric vehicle charging demand has resulted in a higher power load. Accurate charging demand forecasting is an important basis for the realization of refined operation management of charging stations, overall optimized scheduling, and operational strategy upgrades. [0003] Different from traditional power load characteristics, the charging behavior of electric vehicles at charging stations is not only a continuous process in time; due to the mobility of electric vehicles, their charging behavior is closely related to the travel behavior of users. The transfer of will cause the charging demand of different stations to be spatially correlated. Therefore, the charging ...

AI Technical Summary

Problems solved by technology

[0004] On the other hand, due to the lack of accurate charging demand prediction capabilities, although the existing charging capacity management system has certain automatic monitoring and processing methods, its comprehensive dispatching capabilities are limited, and it is only manually adjusted in combination with actual charging use, which is inefficient and the vehicle charging process Contradictions are prominent

Users often face difficulties in finding a car pile during peak hours. Vehicles need to line up for charging and wait for a long time. During off-peak hours, the charging facilities will be idle and the utilization rate is not high. Sometimes, even if the charging facilities are fully charged and output work, due to There is a factor of power demand curve in the charging process. Due to the limitation of charging guns and parking spaces, there will also be a phenomenon that the charging facilities are rich in capacity during peak hours, and the car is forced to wait, that is, the electric vehicle is still charging in the equalizing stage.

Images