基于BP神经网络的CVC冷连轧机板形预测控制模型  

CVC Tandem Mill Flatness Predictive Control Model Based on BP Neural Networks

作　　者：周晓敏[1] 张清东[1] 王长松[1] 陈守群[2] 何汝迎[2] 许健勇[2] 

机构地区：[1]北京科技大学机械工程学院,北京100083 [2]宝山钢铁集团公司冷轧部

出　　处：《北京科技大学学报》2000年第4期374-376,共3页Journal of University of Science and Technology Beijing

基　　金：国家A级课题!"宝钢1420冷连轧机引进技术消化与研究"

摘　　要：将BP神经网络建模方法与预测控制思想相结合用于宽带钢板形自动控制，研究并建立了基于BP神经网络的板形预测控制数学模型．经用宝钢1420mm冷轧实测数据仿真验证表明该模型具有很高的预测精度．The application of BP neural networks and predictive control algorithms to AFC (Auto Flatness Control) system is studied, and a predictive control algorithms based on BP neural networks are offered.Simulation by computer using data get from 1420 CVC tandem mill of Shanghai Baoshan Iron and Steel Company is investigated. The results show that this algorithms can be applicated to AFC system and it's results are good.

关 键 词：BP神经网络 预测控制 CVC 冷连轧机 板形控制 

分 类 号：TG333.11[金属学及工艺—金属压力加工]