基于人工神经网络的扁钢轧制力模型

Modeling of the Rolling Force Based on Artificial Neural Networks

摘要: 根据BP人工神经网络算法原理,结合某厂型钢轧机轧制扁钢时的轧制力实测数据,对扁钢轧制力进行建模.结果表明,神经网络用于轧制力建模是可行的,所建模型系统误差<1%,模型计算值与实测值的偏差<4%,较好地反映了实际轧制过程的特征.

Abstract: Based on the principle of BP neural networks, the rolling force model is created after thoroughly analyzing and Processing the data of 400 nun mill. It states that the difference between the real value and the ours of the model is in order of 4 percent. The model on basis of BPNN is Practical and it reflects the real feature of the rolling process.