基于CGA-SNPOM优化RBF-ARX模型的板形缺陷识别  

Recognition of Flatness Defects Based on RBF-ARX Model Optimized by CGA-SNPOM

作　　者：张秀玲[1,2] 李家欢 魏其珺 董逍鹏 周凯旋 ZHANG Xiu-;LI Jia- huan;WEI Qi- jun;D;ZH(Key Laboratory of Industrial Computer Control Engineering of Hebei Province,Yanshan University,Qinhuangdao 066004,Hebei,China;National Engineering Research Center for Equipment and Technology of Cold Strip Rolling,Yanshan University,Qinhuangdao 066004,Hebei,China)

机构地区：[1]燕山大学河北省工业计算机控制工程河北省重点实验室,河北秦皇岛066004 [2]燕山大学国家冷轧板带装备及工艺工程技术研究中心,河北秦皇岛066004

出　　处：《矿冶工程》2018年第3期127-131,共5页Mining and Metallurgical Engineering

基　　金：河北省自然科学基金-钢铁联合研究基金项目(E2015203354);河北省高校创新团队领军人才培育计划项目(LJRC013);河北省教育厅科学研究计划河北省高等学校自然科学研究重点项目(ZD2016100);秦皇岛市科技局自筹项目(201703A229);2016年燕山大学基础研究专项培育课题(16LGY015)

摘　　要：针对传统优化算法(SNPOM)在辨识RBF-ARX模型参数时易陷入局部最优解的问题,将云遗传算法(CGA)和SNPOM算法结合,提出一种混合优化算法CGA-SNPOM。并以某公司900HC可逆冷轧机板形识别为应用背景,设计了基于CGA-SNPOM优化RBF-ARX的板形缺陷识别模型。分别用SNPOM算法和CGA-SNPOM算法对RBF-ARX模型参数进行优化,仿真验证表明,基于CGA-SNPOM优化的板形识别系统克服了SNPOM容易陷入局部极值的缺点,识别精度大幅提高,是一种有效的板形识别方案。In view of a local optimal solution that is usually obtained by the traditional Structured Nonlinear Parameter Optimization Algorithm（ SNPOM） in identifying parameters of RBF-ARX model,a hybrid optimization algorithm,CGA-SNPOM,was proposed by combining cloud genetic algorithm（ CGA） and SNPOM algorithm. Based on its application into the flatness recognition of a 900 HC reversible cold rolling mill,a pattern recognition model of flatness defects was designed based on RBF-ARX optimized by CGA-SNPOM. The parameters of RBF-ARX were optimized respectively with SNPOM algorithm and CGA-SNPOM,and results show that flatness recognition system based on CGA-SNPOM optimization not only has overcomed the disadvantage of a local extremum led by SNPOM,but also has greatly increased the accuracy of recognition,indicating it is an effective flatness recognition scheme.

关 键 词：板形识别 板形缺陷 SNPOM 云遗传 RBF-ARX 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]