Power distribution network voltage control method based on multi-agent deep reinforcement learning

Abstract

The invention discloses a power distribution network voltage control method based on multi-agent deep reinforcement learning, and relates to the technical field of electrical engineering and the technical field of computer science. The method comprises the steps of: (10) constructing a power distribution network multi-time-scale voltage control model for various voltage regulation devices including a capacitor bank, a photovoltaic inverter and an energy storage device; (20) distributing control variables to a plurality of agents, and converting a voltage control problem into a Markov decision process; (30) adopting a multi-agent deep reinforcement learning algorithm based on a multi-agent depth deterministic strategy gradient to solve the MDP process, and improving the algorithm according to the characteristics of a discrete voltage regulation device; and (40) training and executing the multiple agents to realize the multi-time-scale voltage control method. Compared with the prior art, a multi-time-scale control system is established from the perspective of optimizing the voltage control of the power distribution network, and an algorithm is proposed to process continuous and discrete voltage regulation devices at the same time so as to control the voltage.

Description

technical field [0001] The present invention relates to the field of electrical engineering technology and computer technology, in particular to a distribution network voltage control method based on multi-agent deep reinforcement learning. Background technique [0002] A large number of distributed photovoltaics connected to the distribution network have had a profound impact on the voltage control of the distribution network. At the same time, the development of dynamic reactive power compensation technology, the use of energy storage equipment and controllable distributed energy sources have brought more controllable elements to the voltage control of the distribution network and challenged the existing control methods. [0003] Traditional voltage control methods are mainly based on specific physical models, which are modeled as mixed integer nonlinear programming problems using approximation techniques, and further transformed into various optimization problems. When t...

AI Technical Summary

Problems solved by technology

However, in the existing DRL-based voltage control methods, the action space is often considered to be discrete or continuous, and in the actual distribution network, discrete and continuous voltage regulation devices may exist at the same time; and these methods It is impossible to control different devices in different time scales, and also ignores the important role of the storage system

Images