Electric vehicle load optimization method based on graph convolution and deep belief network

Abstract

The invention discloses an electric vehicle load optimization method based on graph convolution and a deep belief network, and the method comprises the steps: extracting a large amount of uncertain feature data of a road network through a graph convolution network, training the deep belief network through a data driving method, and outputting the optimal charging load distribution of an electric vehicle. The electric vehicle charging load distribution is optimized while the uncertainty of the road network is dealt with, and the influence of the optimization result on the road network and the power grid is quantitatively analyzed. The invention also discloses a load optimization method. The electric vehicle load optimization method is based on a graph convolution and DBN data driving optimization method, uncertain factors influencing the charging load of the electric vehicle can be generalized, and the charging load of the electric vehicle can be accurately and efficiently optimized; compared with solving of a nonlinear optimization model, the method is low in calculation difficulty, high in solving speed and higher in generalization performance for uncertain factors; compared with the existing deep neural network and other technologies, the method has the advantages that the optimization result is more accurate, and the effect of reducing the charging cost of the user is more remarkable.

Description

technical field [0001] The invention relates to the field of power system planning and operation, in particular to an electric vehicle load optimization method based on graph convolution and deep belief network. Background technique [0002] With the depletion of global fossil energy and the increasingly serious environmental pollution problems, electric vehicles have attracted much attention because of their good energy-saving and emission-reduction effects, and their rapid growth in ownership has become an inevitable trend. Electric vehicles have the dual attributes of transportation and electricity consumption, and the charging load of electric vehicles has gradually become an important part of the grid load. However, in the case of a high number of electric vehicles in the future, large-scale disorderly charging will bring a certain burden to the power grid, and a large power impact will lead to problems such as excessive voltage offset and power quality degradation. Sa...

AI Technical Summary

Problems solved by technology

[0003] The existing technology divides the charging load into two categories when regulating the electric vehicle load, one is slow charging at a specific location (such as a family, a parking lot), and is not affected by uncertain factors in the road network; This kind mainly studies the charging load of fast charging stations in cities. Its charging power is high and is affected by various uncertain factors such as charging price, traffic flow, and charging queuing time. It has strong randomness and uncertainty. How to accurately evaluate it? And optimizing the charging load of this type of electric vehicle is the most difficult point of research

[0004] At present, most of the relevant studies model the driving and charging behavior of a large number of electric vehicles as a Nash equilibrium (User equilibrium, UE). Generalization performance, as the scene changes, a large number of optimizations need to be re-solved, which increases the computational burden

Images