Optimization For The Cold Tandem Rolling Schedule Based On SA-PSO

Iron and steel industry is an important energy industry for an industrialized country. For a long time, the ratio of rolling strip production is an important symbol that measuring the production level of its metallurgy industry, especially for the production of rolling strip steel and the level of its making technology, that are very significant for a country to its development in the key field of technology. Along with the advancing of the quality of cold rolling products for market demand and for promoting of production efficiency, every company of iron & steel is urgent to improve its level of production and technology. Of that, the rolling schedule is one of main study object in cole-rolled strip steel production technology, and also it is the core element of pre-setting for rolling parameters of rolling math models. It is cannot fit the needs of cold rolling with production experiences for pre-setting rolling schedule in the old time. Therefore, it is significance to study one rolling schedule that fits modern rolling production.Firstly, the models of main parameters in the pr-setting rolling schedule were studied and analyzed. For the control system of cold rolling, the work of the cold rolling was analyzed in detail. And with the real producing, the parameter computing models for designing rolling schedule was deduced in thesis. For the linkage of the rolling force models in the system, and to prevent the intricate computing of this model or bad precision made by this linkage, the algorithm of BP NN connected with the classical model was designed to preset the rolling force. Secondly, a new model of slipping factor was studied and the model of optimizing rolling schedule was designed. Base on the analyzing for the work of slipping factor producing in the rolling, and with the real causes of slipping phenomenon, a new model of slipping factor was deduced, the two models were compared with their nature and precision. In the meantime, some key factors that affect the slipping factor were analyzed. After that, base on the fully analyzing to the old setting model of rolling schedule, and for the lack of connecting with slipping factor in rolling schedule setting in old system, the sequential optimization model of rolling schedule setting was deduced with the aim function of equal measure forcing down. Lastly, the SA-PSO was be used to optimize the rolling schedule. Through analyzing the simulated annealing algorithm and the Particle Swarm Optimization algorithm, the Metropolis criterion of the first algorithm was introduced to the later, this new algorithm could avoid precocious convergence in the training just like PSO. This is the reformed PSO based on the SA, and it has been used in the designing of optimizing rolling schedule.The simulation and analysis for rolling schedule optimization had been made with the real data by MATLAB software in the thesis for the model of roll schedule optimization. And the results verified the feasibility and validity of the schedule model. Just also, it shown that the SA-PSO algorithm strengthened the swarm’s ability to get rid of the partied optimal solution, and had higher convergence rate and precision.