摘要
在实际生产过程中 ,传统轧制力模型在计算中板轧机轧制力时存在着较大的误差 ,为了提高中厚板轧机轧制力的预报精度 ,提出了一种将人工神经元网络用于轧制力预报数学模型中 ,进行轧制力预设定。离线仿真表明 ,采用本文所述的方法 ,预报精度要优于传统的数学模型 ,预报精度的相对误差可以控制在± 4
In the process of real production, traditional models have much limitation to calculate the rolling force. In order to improve the prediction precision of the rolling force, a new method using artificial neural networks in prediction models of rolling force is put forward. Off line simulation indicates that the predicted results are more accurate than the one estimated with traditional models. The relative error is within ±4%.
出处
《材料与冶金学报》
CAS
2002年第2期150-153,共4页
Journal of Materials and Metallurgy
基金
国家自然科学基金资助项目 (No .5 0 10 40 0 4)
