Evolutionary Algorithm In Solving Multi-objective Optimization Problems

Evolutionary algorithm in solving multi-objective optimization problems and application in power systems is a hot topic in the field of algorithms.With the goal of optimizing the problem is getting higher and higher,more and more goals,The algorithm also becomes complicated.The traditional genetic particle swarm,differential evolution and other algorithms have been unable to effectively solve such problems.With the study,multi-objective evolutionary algorithm has a good effect in solving high-dimensional multi-objective optimization problems.This paper is based on the traditional optimization algorithm,introduces the multi-objective particle swarm optimization algorithm(MOPSO)to solve the multi-objective optimization problem and introduced the application of improved differential evolution algorithm in Solving the Environment / Economic Dispatch(EED)of Power System.The details are as follows:Firstly,introduce the concept of Particle Swarm Optimization(PSO)in detail and a variety of improved algorithms based on PSO.Compare the effects of various improved PSO algorithms in standard test functions and analyze the cause.Which provides a theoretical basis for further study of MOPSO algorithm.Followed by the use of Particle Swarm Optimization(PSO)prior knowledge,A MOPSO algorithm which add external archives and local perturbation strategies is proposed because of the traditional PSO algorithm can not effectively solve the multi-objective optimization problem.And proved by experimental results,This paper describes the MOPSO than traditional NSGA2 can get a higher degree of non-dominated solution set and better Pareto frontier.Then the basic concepts of differential evolution algorithm(DE)are described in detail,list a variety of extended modes and compare them.At the same time,introduce a variety of improved DE algorithm,and comparison in the standard test function.This chapter lay the theoretical foundation for the post-research DE algorithm in power system EED application.Finally,the power system EED problem is introduced,This is a high-dimensional,multi-objective,multi-constrained optimization problem,Traditional algorithms can not be solved effectively.To this end,the use of DE algorithm a priori knowledge,The problem of two-objective optimization is transformed into single-objective optimization problem by penalty factor,use heuristic strategies to resolve multiple constraints and join the priority list method.The experimental results show that the algorithm proposed in this paper can make the energy consumption of the generator with higher output as much as possible,So as to get a better solution.