Cluster electric vehicle charging behavior optimization method based on deep reinforcement learning

Abstract

The invention discloses a cluster electric vehicle charging behavior optimization method based on deep reinforcement learning, and belongs to the technical field of new energy vehicle optimization management. The invention relates to a deterministic strategy gradient algorithm based on double delay depths. Modeling is carried out on the power continuous adjustable charging process of the electricvehicle, an intelligent agent is trained to control the charging power, the charging behavior of the electric vehicle is optimized, the load when the time-of-use electricity price is high is transferred to the load when the electricity price is low, and the purposes of reducing the user charging expenditure and stabilizing the peak-time load of a power grid are achieved. Compared with a traditional optimization control method, the TD3 has obvious advantages in speed and flexibility, and the problems that an existing reinforcement learning method is discrete in action space, difficult in training convergence and poor in stability can be effectively solved. In order to enhance the generalization ability of the intelligent agent, noise is added on the basis of original state observation, a group of electric vehicles with different initial SOCs and different arrival and departure time are simulated, and the method is expanded to cluster electric vehicle charging behavior control.

Description

technical field [0001] The invention belongs to the field of power system optimization scheduling, in particular to a method for optimizing charging behavior of cluster electric vehicles based on deep reinforcement learning. Background technique [0002] China attaches great importance to the development of the new energy vehicle industry; it is estimated that by 2025, the sales volume of new energy vehicles will account for about 25%. Reached about 7 million. The high power and space-time uncertainty of electric vehicles will change the existing load level of the power grid, further increase the peak-valley difference, and have an impact on the safety and stability of the power grid. [0003] As an important means of demand-side management, peak-valley time-of-use electricity pricing plays an important role in guiding and regulating electricity consumption behavior and assisting power grids in peak-shaving and valley-filling. The load aggregator (aggregator) can respond t...

AI Technical Summary

Problems solved by technology

Compared with the traditional optimization control method, TD3 has obvious advantages in speed and flexibility, and can effectively overcome the problems of discrete action space, difficult training convergence and poor stability of previous reinforcement learning methods

Images