Research On Mopso Based On Integrating Multiply Strategies And Application

The optimization of strip cold rolling process can be regarded as a nonlinear constrained multi-objective optimization problem.With the development of science and technology,the quality of product in people's life is getting higher and higher,so it is more and more fine to regulate it.Based on the 1250 mm five-stand tandem mill plant as the research background,by improving the traditional multi-objective particle swarm optimization(PSO)algorithm,is proposed a fusion strategy more multi-objective particle swarm optimization algorithm to optimize the rolling regulations formulated.The main contents of this paper are as follows.In this paper,three improvement strategies are proposed to solve the problem of multi-objective optimization algorithm,which is easy to get into local optimal solution and poor distribution.First of all,the traditional particle swarm optimization algorithm,rely mainly on self of particles in the population information and share information among groups of population to select the optimum solution,to obtain the solution set of distribution difference,and the convergence speed is too slow.Second,the particle swarm algorithm easy to fall into the predicament of the local optimum,populations can no longer for the better evolution,and the appropriate variable disturbance is conducive to jump out of local optimum in time of particles in the population,so in the introduction of a multi-criteria variation of particle swarm optimization(PSO),speed up the convergence speed of the algorithm;Finally,in the process of iteration algorithm,the parent,child and external archive three populations were non dominated sorting,select high level solution as elite solution in the external file,and through the traditional external archive maintenance mechanism,the external archive in the solution set for maintenance,ensure uniform output distribution of the final solution,has a good variety.In order to verify the validity of the MSIMOPSO,uses the ZDT,DTLZ and some of the UF test function set simulation experiment,the results show that through the decomposition and non-dominated sorting combination,as well as the effective dynamic variation in the process of evolution,the overall performance of algorithm has great improvement.Based on the research background of 1250 mm five-frame cold tandem mill,the effective constraint restriction was established by analyzing the rolling process and the parameter performance of the equipment.On this basis,the three-objective optimization model is established to prevent skidding,etc.By improving the algorithm MSIMOPSO,the simulation results show that by optimizing,reducing the sliding factor,balancing the load of each frame,and improving the quality of the final frame plate.