[1]张秀玲,陈丽杰,逄宗朋,等.RBF神经网络的板形预测控制[J].智能系统学报,2010,5(1):70-73.
ZHANG Xiu-ling,CHEN Li-jie,PANG Zong-peng,et al.A predictive system for process control of flatness in rolling mills using a radial basis function network[J].CAAI Transactions on Intelligent Systems,2010,5(1):70-73.
点击复制
ZHANG Xiu-ling,CHEN Li-jie,PANG Zong-peng,et al.A predictive system for process control of flatness in rolling mills using a radial basis function network[J].CAAI Transactions on Intelligent Systems,2010,5(1):70-73.
RBF神经网络的板形预测控制
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
5
期数:
2010年第1期
页码:
70-73
栏目:
学术论文—机器学习
出版日期:
2010-02-25
- Title:
- A predictive system for process control of flatness in rolling mills using a radial basis function network
- 文章编号:
- 1673-4785(2010)01-0070-04
- Keywords:
- shape control; HCmill; hydraulic control of bending rollers; RBF neural network; predictive control
- 分类号:
- TP18
- 文献标志码:
- A
- 摘要:
- 由于板带轧制的环境十分复杂，如温度的变化是无法避免的干扰，以及HC轧机液压弯辊系统的非线性和不确定性，使得按传统理论建立的模型和控制方法都难以达到理想的效果.针对这一问题，提出了一种基于径向基函数(RBF)神经网络的模型预测控制方案应用于带材控制中，以提高带材的成材率，充分发挥液压弯辊力对板形的调整作用，改善轧机系统的动态特性.仿真结果表明了该控制系统的性能良好，有较强的抗干扰能力和较好的鲁棒性和快速性.
- Abstract:
- When plate and strip rolling is done in very complex environments, such as high crown (HC) rolling mills, there are many factors that make system control difficult. Factors affecting the flatness of steel sheets include temperature changes as well as nonlinearities that lead to uncertainty about results from bending roller forces. A novel predictive control program was proposed, one employing a radial basis function (RBF) neural network. It ensures flatness by controlling the bending forces of rollers. Simulation results confirmed this scheme has good performance and robustness.
- 参考文献/References:
-
［1］宗晓萍,冯贺平. 基于神经网络的时滞系统预测控制［J］. 控制理论与应用, 2005, 24(12): 68.
ZONG Xiaoping, FENG heping. Neural networkbased predictive control for timedelay systems［J］. Control Theory and Application, 2005, 24(12): 68.
［2］HARTER F P, De CAMPOS VELHO H F. New approach to applying neural network in nonlinear dynamic model［J］. Applied Mathematical Modelling, 2008, 32(12): 26212633.
［3］CHOI J N, OH S K, PEDRYCZ W. Identification of fuzzy models using a successive tuning method with a variant identification ratio［J］. Fuzzy Sets and Systems, 2008, 159(21): 28732889.
［4］邓继雄,李志舜,梁 红.确定神经网络参数的新方法［J］. 微处理机, 2006, 27(4): 4849.
DENG Jixiong, LI Zhishun, LIANG Hong.A new method of ascertaining radial basis function network parameter［J］. Microprocessors, 2006, 27(4): 4849.
［5］张秀玲.冷带轧机板形智能识别与智能控制研究［D］. 秦皇岛: 燕山大学, 2002.
ZHANG Xiuling. Research on intelligent control and recognition of flatness for cold strip mill［D］. Qinhuangdao: Yanshan University, 2002.
［6］连家创,刘宏民. 板形板厚控制［M］. 北京： 兵器工业出版社, 1996.
［7］张秀玲，刘宏民.神经网络模型参考自适应控制及其在带材板形控制系统中的应用［J］.机械工程学报， 2001, 37(9): 8387.
ZHANG Xiuling, LIU Hongmin. Neural network model reference adaptive control and its application in the strip shape control system［J］. Chinese Journal of Mechanical Engineering, 2001, 37(9): 8387.
［8］陆冬娜,杨马英. 基于RBF神经网络的非线性模型预测控制［J］. 浙江工业大学学报, 2007, 35(2): 123126.
LU Dongna, YANG Maying. Nonlinear model predictive control based on RBF networks［J］. Journal of Zhejiang University of Technology, 2007, 35(2): 123126.
- 相似文献/References:
-
[1]张秀玲,逄宗鹏,李少清,等.ANFIS的板形控制动态影响矩阵方法[J].智能系统学报,2010,5(4):360.
ZHANG Xiu-ling,PANG Zong-peng,LI Shao-qing,et al.A dynamic influence matrix method for flatness control based on adaptivenetworkbased fuzzy inference systems[J].CAAI Transactions on Intelligent Systems,2010,5():360.
备注/Memo
收稿日期：2008-10-06.
基金项目：国家自然科学基金资助项目（50675186）.
通信作者：张秀玲.E-mail:zxlysu@yahoo.com.cn.
作者简介：
张秀玲， 女，1968年生，教授、博士，主要研究方向为神经网络智能控制.获国家科技进步二等奖1项，省部级科技进步一等奖、二等奖各1项，发表学术论文60余篇.
陈丽杰，女，1982年生,硕士研究生，主要研究方向为神经网络优化板形控制系统设计.
逄宗朋，男，1983年生，硕士研究生，主要研究方向为模糊神经网络优化板形设计.
更新日期/Last Update:
2010-04-06