Improvement Of Multi-objective Differential Evolution Algorithm And Its Application In Rolling Schedule Optimization

With the continuous development of society,the scale and complexity of optimization problems have gradually increased.Traditional methods have been difficult to solve optimization problems,and intelligent optimization algorithms have emerged.Differential evolution algorithm has received extensive attention due to its simple structure,good convergence performance,few control parameters,and easy implementation.However,the differential evolution algorithm still has shortcomings such as easy to fall into the local optimal,search stagnation,etc.,which limit its performance and hinder the promotion of its application.Therefore,the research on the improvement of differential evolution algorithm has important theoretical research significance and practical application value.Aiming at the above shortcomings,this paper has made some improvements to the multi-objective differential evolution algorithm and applied it to the optimization of the rolling schedule of 1250 mm tandem cold rolling in a steel mill.The main contents are as follows:(1)Aiming at the deficiencies in convergence and diversity of differential evolution algorithm when dealing with complex multi-objective optimization problems,a multi-objective differential evolution algorithm based on two-stage diversity enhancement(TSDE)is proposed.In order to select high-quality parents and produce a better-performing offspring population,the first stage adopts the optimized cell density method.This method can estimate the global distribution of the target space,identify dense areas,and increase the diversity of the initial population of each generation.The second stage is to introduce the principal component analysis(PCA)operator into the external archive to perturb the non-dominated solutions,and generate a set of new solutions around the original non-dominated solution set,so as to improve the diversity while ensuring the convergence.(2)In order to balance global exploration and local exploitation,an adaptive multi-objective differential evolution algorithm(AMODE-MPS)based on mirror point screening strategy is proposed.Aiming at the problem of poor quality of the parent population,the mirror point screening method is applied to the generation process of the initial population to improve the diversity of the parent population.In order to solve the problem of imbalance between exploration and development,adaptive chaotic mutation is introduced in the exploration process to help the population jump out of the local optimum.(3)In order to improve product quality and reduce cost consumption,this paper uses an improved differential evolution algorithm to optimize rolling schedules.Taking equal load and preventing slipping as the objective function,according to the production purpose and production characteristics,a multi-objective optimization model of rolling schedule is established.