锻压技术

隔热屏辊轧机板形控制系统研究及实验分析

英文标题：Research and experimental analysis on plate shape control system of heat shield rolling mill

作者：马玉春

分类号：TP391.9

出版年,卷(期):页码：2024,49(11):70-76

摘要：

针对具有非线性、时变性的隔热屏辊轧机液压辊轧力控制系统，设计了一种基于模糊神经网络模型参考的自适应控制器，融合了模糊控制和神经网络的长处，在解决不确定性等问题上具有优势。该自适应控制器通过调整串并联模型辨识器的权值，以及采用附加动量项的梯度下降法训练网络，实现了对系统的自适应控制。对该控制器进行仿真分析，结果表明：该控制器具有较强的抗干扰和信号追踪能力，能准确跟踪辊轧力的目标设定值。最后，构建了隔热屏辊轧机下压辊调整控制及检测系统结构，并将该控制器应用于此系统进行实验验证，结果表明：薄壁筒件样件的实测辊弯尺寸相比设计值的偏差在0.3%以内，具有较高的可靠性和精度。

For the hydraulic rolling force control system of heat shield rolling mill with the characteristics of nonlinearity and time-varying, a self-adaptive controller based on fuzzy neural network model reference was designed, which combined with the advantages of fuzzy control and neural network and had advantages in solving uncertainty problems. Then, the self-adaptive controller realized the self-adaptive control of the system by adjusting the weights of series-parallel model recognizer and training the network by gradient descent method with additional momentum term. Furthermore, the simulation analysis on the controller was carried out, and the results show that the controller has strong anti-interference and signal tracking ability, which could accurately track the target setting value of rolling force. Finally, the structure of lower press roller adjustment control and detection system for heat shield rolling mill was constructed, and the controller was applied to this system for experimental verification. The results show that the measured roll bending size of thin-walled cylinder sample has a deviation of less than 0.3% compared to the design value, which has high reliability and accuracy.

基金项目：

国家自然科学基金资助项目（51774162）

作者简介：马玉春（1993-），男，硕士，工程师 E-mail：glove0908@163.com

参考文献：

[1]王晓晨，杨荃，孙友昭. 六辊冷轧机的弯辊力组合板形控制策略[J]. 工程科学学报，2014，36(6)：824-829.

Wang X C, Yang Q, Sun Y Z. Shape control strategy by bending force combination adjustment for 6-hi cold rolling mill[J]. Chinese Journal of Engineering,2014,36(6):824-829.

[2]于仲安，杨国亮. 基于大脑情感学习模型的参考自适应液压伺服控制系统研究[J]. 计算机测量与控制，2013，21(3)：671-673，680.

Yu Z A, Yang G L. Research of model reference adaptive hydraulic servo control system based on brain emotional learning model[J]. Computer Measurement & Control,2013,21(3):671-673，680.

[3]陈志旺，李亮. 模糊模型参考自适应轧机位置控制[J]. 冶金设备，2008(1)：6-9，5.

Chen Z W, Li L. Fuzzy MRAC for position control of rolling mill[J]. Metallurgical Equipment,2008(1):6-9，5.

[4]陈丽，李建更，乔俊飞，等. FNN模型参考自适应控制在轧机液压弯辊系统中的应用[J]. 计算机测量与控制，2005(12)：1396-1399.

Chen L, Li J G, Qiao J F, et al. Fuzzy neural network model reference adaptive control method for the hydraulic bending roll in the control system of UC rolling mill[J]. Computer Measurement & Control, 2005(12):1396-1399.

[5]吕其栋. 冷连轧机板形智能控制策略的研究[D]. 沈阳：东北大学，2008.

Lyu Q D. Research on Intelligent Control Strategy of Flatness for Tandem Cold Mill[D]. Shenyang: Northeastern University, 2008.

[6]Pin G, Francesconi V, Cuzzola F A, et al.Adaptive task-space metal strip-flatness control in cold multi-roll mill stands[J]. Journal of Process Control, 2013,23 (2):108-119.

[7]Jia C Y, Bai T, Shan X Y, et al. Cloud neural fuzzy PID hybrid integrated algorithm of flatness control[J]. Journal of Iron and Steel Research(International),2014,21(6):559-564.

[8]王海峰. 冷轧带材板形控制系统模型优化及仿真软件开发[D]. 秦皇岛：燕山大学，2020.

Wang H F. Model Optimization and Simulation Software Development of Flatness Control System for Cold Rolled Strip[D]. Qinhuangdao: Yanshan University, 2020.

[9]白涛. 基于遗传算法优化的云模型PID液压弯辊板形控制[J]. 塑性工程学报，2021，28 (4)：206-211.

Bai T．Hydraulic bending roll flatness control of cloud model PID based on genetic alogorithm optimization[J]. Journal of Plasticity Engineering, 2021,28 (4):206-211.

[10]孙卫华. 轧机液压AGC系统数学模型及控制方法研究[D]. 沈阳：东北大学，2008.

Sun W H. The Study on the Dynamic Model and Control Method of Hydraulic AGC System[D]. Shenyang: Northeastern University, 2008.

[11]陈丽.模糊神经网络及其在板形控制中的应用[D]. 北京:北京工业大学，2005.

Chen L. Fuzzy Neural Network and Its Application to Shape Control[D]. Beijing:Beijing University of Technology,2005.

[12]宋晓东，姚晓先，宫展博. 调压型智能泵源的模型参考自适应控制[J]. 北京理工大学学报，2011，31(8)：944-948.

Song X D, Yao X X, Gong Z B. Model reference adaptive control for variable-pressure pumping source[J]. Transactions of Beijing Institute of Technology,2011,31(8):944-948.

[13]马晨龙. 基于模型参考自适应控制的列车速度控制研究[D]. 兰州: 兰州交通大学，2023.

Ma C L. Research on Train Speed Control Based Adaptive Control on Model Reference[D]. Lanzhou: Lanzhou Jiaotong University, 2023.

[14]张敏，徐启华. 基于改进BP的神经网络模型参考自适应控制[J]. 软件，2015，36(7)：118-123.

Zhang M, Xu Q H. The improved BP neural network model reference adaptive control[J]. Computer Engineering & Software,2015,36(7):118-123.

[15]Patino H D, Liu D R. Neural network-based model reference adaptive control system[J]. IEEE Transactions on Cybernetics, 2000, 30(1):198-204.

[16]Da F P, Song W Z. Fuzzy neural networks for direct adaptive control[J]. IEEE Transactions on Industrial Electronics, 2003, 50(3):507-513.

[17]张秀玲. 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)：83-87.

Zhang X L. Neural network model reference adaptive control and its application in the strip shape control system[J].Chinese Journal of Mechanical Engineering，2001, 37(9):83-87.

[18]楼顺天，胡昌华，张伟,等. 基于Matlab的系统分析与设计—模糊系统[M]. 西安:西安电子科技大学出版社，2001.

Lou S T, Hu C H, Zhang W, et al. System Analysis and Design Based on Matlab—Fuzzy System[M]. Xi′an: Xidian University Press,2001.

[19]孟劦，胡亚洲，陈晓. 基于自适应模糊神经网络控制器的网络控制系统[J]. 计算机工程与应用，2015，51(18)：90-93,208.

Meng J, Hu Y Z, Chen X. Network control system based on adaptive fuzzy neural network controller[J]. Computer Engineering and Applications,2015,51(18):90-93,208.

[20]陈伯芳，尹平林，马龙. 基于模糊神经网络的温度控制系统研究[J]. 计算机与数字工程，2010，38 (7)：54-57.

Chen B F, Yin P L, Ma L. Research on fuzzy neural networks for temperature control system[J]. Computer & Digital Engineering, 2010, 38 (7): 54-57.

[21]徐国凯，张涛，王娟,等. 自动控制原理[M]. 第3版.北京:清华大学出版社，2017.

Xiu G K, Zhang T, Wang J, et al. Principles of Automatic Control[M].Third Edition. Beijing: Tsinghua University Press,2017.

[22]HB 5936—2011,零组件标记[S].

HB 5936—2011, Marking of parts and assemble[S].

服务与反馈：

【文章下载】【加入收藏】