Optimization Of Aluminum Hot Rolling Schedule Based On Improved Artificial Fish Swarm Algorithm

For decade years, industries of global aluminum had entered a new developmentstage. The aluminum processing industry developed fast and became ones of the pillarindustries of many countries. Aluminum hot tandem rolling is a complex system process,and establishing a reasonable rolling schedule is not only an important condition ofguaranteeing the high quality, high efficiency and low energy consumption of the hotrolling mill operation, but also is an important technology issue of comprehensiveoptimization in hot rolling production. In recent years, with the rapid development ofindustrial production, the quality and production of aluminum plate are both requiredincreasingly high. Therefore, higher demands for the optimization ability of rollingschedule are put forward. With the development of the computer technology, a newchapter for the development of the rolling theory was opened by the application ofartificial intelligence in rolling fields. Compared with the traditional rolling methods, theoptimal control for the rolling process was achieved by the artificial intelligencealgorithms which avoid endless search for the deep laws of rolling process in the past andsimulate human brains to deal with the practical process on the basis of the fact and data.The experience and summary in practical production can be raised to the theory level byformulating the rolling schedule. Therefore, artificial intelligence algorithm has a moreextensive significance.As a new intelligent optimization algorithm, artificial fish swarm algorithm has agood ability to overcome the local extreme and obtain the global extreme, it only uses thetarget function value, the realization of the algorithm is simple and easy to understand.Currently, artificial fish swarm algorithm has been applied to many practical areas, andgood effect has been shown in many complex optimization problems. However, like othergroup intelligence evolutionary algorithm, artificial fish swarm algorithm also has theshortcomings such as premature and slow convergence speed. So in the basic fishalgorithm is improved in this paper and the improved on the basis of the algorithm hasbetter optimization effect. This paper took one factory of aluminum strip finishing rolling mill in Henan as theresearch object. Accorded to lots of actual measured data, regressed the deformation stressmodel of different alloy. Considering various factors which lead to slip phenomenon, theconcept of hot rolling slip factor is proposed and a multi-objective function with equalpower margin and preventing the slip phenomenon is established, then the improvedartificial fish swarm algorithm is applied to optimize the reduction. The simulationpractice shows that the satisfactory results can be obtained by the optimal rolling schedule.