Electric vehicle charging demand prediction method, system and device based on neural network

Abstract

The invention belongs to the field of electric vehicle charging demand prediction, particularly relates to an electric vehicle charging demand prediction method, system and device based on a neural network, and aims to solve the problem that an existing charging demand prediction method is difficult to realize accurate and precise charging demand prediction. The method comprises the steps: dividing a to-be-predicted area into grids of different sizes, obtaining charging pile information and charging demands in a set historical time period as model training data, and obtaining weather forecastinformation and holiday information as model auxiliary training data; sampling and normalizing the data; constructing a charging demand prediction model based on the neural network, and setting an activation function, a loss function and a cost function of the model; performing model training through the training data and the auxiliary training data; performing charging demand prediction through the model obtained by training. The weather information and holiday information are used for assisting model training, and prediction result accuracy is high.

Description

technical field [0001] The invention belongs to the field of electric vehicle charging demand prediction, and in particular relates to a neural network-based electric vehicle charging demand prediction method, system and device. Background technique [0002] The transformation of the automobile industry to electrification has become an important trend in the sustainable development of society. With its good environmental protection and energy-saving advantages, electric vehicles have become an important direction for the future development of the automobile industry. It can be seen that the scale of electric vehicles in my country will be further expanded in the future. [0003] However, the contradiction between the charging demand of electric vehicles and charging stations has become increasingly prominent. Among them, the accurate prediction of electric vehicle charging demand is the most basic management method in the management of electric vehicle charging network, and...

AI Technical Summary

Problems solved by technology

However, most of the researches use ANN, SVM, KNN and other shallow models to make one-dimensional time series prediction, and the model learning ability and expression ability for complex systems are limited.

[0005] With the further development of mobile charging technology, the number of mobile charging piles has further increased. When people predict the charging demand of electric vehicles, they will not only consider whether charging can be realized, how to reduce the impact on the power grid and reduce charging costs, etc., but also further Taking into account factors such as weather conditions and the distance between the charging pile and the car to be charged, which further affects the accuracy of the electric vehicle charging demand forecast

[0006] In general, it is difficult to establish an accurate probability model for traditional electric vehicle charging demand forecasting methods, and can only achieve simple estimation of electric vehicle charging demand, while the existing intelligent forecasting methods based on machine learning do not consider the influence of weather and charging distance The impact of charging demand, so the prediction accuracy of electric vehicle charging demand is low

Images