基于自适应遗传算法的冷连轧负荷分配优化

Load Distribution Optimization in Tandem Cold Rolling Based on Adaptive Genetic Algorithm

针对冷连轧轧制力模型精度低的问题,利用BP神经网络预测变形抗力和摩擦因数,并与现有的轧制力解析模型相结合,用来提高轧制力的设定精度,再以各机架轧制力等比例均衡分配为优化目标,采用改进的自适应遗传算法,设计了一种冷连轧负荷分配优化方法。通过对某五机架冷连轧机的负荷分配进行比较,结果表明自适应遗传算法具有比标准的遗传算法收敛性能更好、精度更高等优点,可以作为冷连轧负荷分配优化的新方法加以推广。

In view of imperfection of the rolling force model in tandem cold rolling, the paper use BP neural network to predict the deformation resistance and friction coefficient then combining with the mathematical model in order to improve the precision of the model. A load distribution method was designed with improving adaptive genetic algorithm in which rolling pressure relative balance was the optimized objective function. The load distribution method was compared with the existing method on five--stand tandem cold rolling mills. Experiments results demonstrate that the improved adaptive genetic optimization algorithm have the advantages of better convergence and higher precision. It provides an effective method in the intelligent optimization for tandem cold milling.