Optimization Of Aluminum Strip Multi-pass Hot Rolling Schedule And Development Of Software

Abstract:Setting-up calculation of rolling schedule is the core issue of the aluminum strip hot rolling production process. Reasonable rolling schedule can not only make the full use of rolling mills and improve their efficiency, but also eliminate the difficulty of follow-up control and improve quality of the final product. With the development of the optimization algorithm and computer technology, the practical application of genetic algorithm and its improved algorithms has been more widely used. Therefore, it is of great realistic significance to the production of the aluminum strip hot rolling process by effectively improving genetic algorithm and optimizing the rolling schedule.In setting-up rolling schedule, the most important two types of indices are energy consumption and the quality of strip shape. A reversible four-high rolling mill is studied in this paper. And according to the aluminum strip hot rolling process features, combining with the classical theory model, firstly, the relevant mathematical models for rolling process, including rolling force model and convex model, were determined, and then the prediction accuracy of models was verified.Secondly, many factors which affect strip shape, were analyzed by means of the method of theoretical analysis combining with mathematical regression, and flatness model was established. With the finite element method of the MARC simulation software, the aluminum strip rolling process was simulated and the influence factors of strip flatness of multi-pass hot rolling were analyzed and the prediction accuracy of the flatness model was verified.Then the genetic algorithm was improved, such as improvement of the crossover and mutation probabilities as the adaptive change of the evolutional generation, and optimization of the selection operation and penalty function, which guarantees the whole search scopes, and improvement of the convergence speed of the algorithm. Comparing the practical example by optimizing the aluminum strip hot rolling process with production data, the result shows that the optimization results, whether single objective optimization or multi-objective optimization, are better than production data, which will be a good reference to the aluminum strip hot rolling production.At last, a rolling schedule optimization offline software was been developed with C#on VS2010development platform. The interface of the software is terse, and the system is easy to operate.38figures,12tables,60references.