Intelligent PI control method for dual-active full-bridge converter

Abstract

The invention discloses an intelligent PI control method for a dual-active full-bridge converter, and the method comprises the steps: adding a parameter regulator based on deep reinforcement learning in PI control, so as to intelligently adjust the parameters of PI control, thereby achieving the self-adaption of a complex operation environment, and improving the dynamic performance of the dual-active full-bridge converter. A TD3 algorithm is adopted, an action value A (dkp (t) and dki (t)) is generated through an action network, then the award r of the action is judged through the award mean value of two judgment networks, and the problems of overestimation behavior and instability of a single network judgment value are avoided. And finally, PI control parameters are intelligently adjusted so as to adapt to a complex operation environment, and an output voltage error is minimized, so that good dynamic performance is obtained. According to the invention, the problems of weak dynamic regulation capability and to-be-improved dynamic performance caused by fixed parameters and incapability of self-adaptive regulation of a traditional dual-active full-bridge converter PI control method are solved.

Description

technical field [0001] The invention relates to the technical field of dual-active full-bridge converters, in particular to an intelligent PI control method for dual-active full-bridge converters using deep reinforcement learning. Background technique [0002] With the increasing depletion of fossil energy and the more serious environmental problems it brings, renewable new energy has received extensive attention. Because it is mostly in the form of DC and the voltage needs to be transformed, it requires a large number of power electronic converters. With the rapid development of switching device technologies such as MOSFET and IGBT, dual active full-bridge converters have attracted more and more attention due to their simple structure, bidirectional energy transmission, high power supply quality, and flexible control methods. [0003] At the same time, in order to adapt to the access of distributed renewable energy and various loads to the power grid, the dual active full-b...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problem that the PI control parameters of the traditional dual-active full-bridge converter cannot be adaptively adjusted, and aiming at the defects of the existing PI control strategy, the present invention proposes an intelligent PI for the dual-active full-bridge converter based on deep reinforcement learning. Control Method

Images