基于布谷鸟算法的铝热连轧轧制规程优化

Optimization of Aluminum Hot Tandem Rolling Schedule Based on Cuckoo Search Algorithm

根据轧制原理,提出了支持向量机建立轧制力预报模型,并通过布谷鸟算法优化支持向量机参数,达到提高预报精度的目的。提出了支持向量机网络与数学模型结合的方法,对某"1+4"铝热连轧厂现场采集的5052铝合金轧制数据进行离线仿真,进一步提高了轧制力预报精度。在轧制规程设定中,建立了以预防打滑为主的电机功率剩余程度相近目标函数,并用布谷鸟算法对压下率进行优化,结果表明,该规程有很好的预防打滑效果,并能保证各机架电机的功率剩余程度相近。

According to the principle of rolling, a forecasting model for roiling force based on support vector machine （SVM） was proposed. In order to improve the prediction precision, the cuckoo search algorithm was used to optimize the parameters of SVM. The combination of SVM network with mathematic model was proposed. The offline simulation of 5052 aluminum alloy roiling data obtained on site of a ＂1 ＋4＂ aluminum strip rolling factory was carried out to further improve the precision of roiling force prediction. The objective function based on the prevention of slip and residual power of motors was established in the specification of rolling and the rolling reduction was optimized by cuckoo algorithm. The results show that this method has good effects of preventing slip and can ensure similar residual power of each motor.