Self-learning optimization control of microgrid electric energy based on adaptive dynamic programming method
WEI Qinglai
Abstract：[Objective]The optimization of electricity supply and demand matching and regulation in smart grids is becoming increasingly complex,and traditional static optimization methods cannot meet the optimization requirements of smart grids.To this end,a self-learning optimal control method was proposed to solve the optimal control problem of ice storage air conditioning(IAC)systems.[Methods]The adaptive dynamic programming-particle swarm optimization(ADP-PSO)algorithm was adopted to address the optimal control problem of the systems.A two-layer iterative adaptive dynamic programming method was designed to learn the optimal control strategy,where the inner iteration calculated the sequence of transformed iterative control laws,and the outer iteration optimized the iterative value function.Meanwhile,a parallel control scheme was developed to obtain the optimal control suitable for the IAC system,which could meet the cooling demand at the lowest operating cost.[Results]Simulation results and comparative studies verify the effectiveness of the proposed algorithm.[Conclusions]The proposed ADP-PSO algorithm can achieve optimal energy matching.This strategy can make the iterative value function converge to the optimum,thereby obtaining the optimal control strategy and minimizing the system operating cost.
Keywords：smart gridadaptive dynamic programmingparticle swarm optimizationdata-based controlneural network
Publication Date：2025-11-25
Online Publishing Date：2025-12-29(First online date of this platform, not the publication date of the document)
Pages：7( 681-687 )