针对目前热轧铝板带凸度控制存在的问题,建立以轧辊温度在线测量为基础的分段冷却闭环模糊控制系统。以简单的测量设备和控制方法,代替昂贵复杂的板带凸度控制机构。用实际轧制数据训练自适应PSO-BP神经网络,并用训练完成的神经网络依据目标板带凸度得出轧辊温度预设定模型;依据操作人员的经验以及理论分析结果,设计分段冷却模糊控制规则,形成分段冷却闭环控制系统,达到控制板带凸度的目的。经在某厂二辊可逆热轧上的应用,结果表明：轧辊温度偏差量可控制在±4 ℃内;铝板带纵向各处的凸度95%以上可控制在目标凸度(20~40 μm)范围内。该方法充分发挥了分段冷却系统对板带凸度的控制能力。
关键词：
热轧; 铝合金; 凸度; 控制; 分段冷却
Based on measuring system of the roll temperature, the closed-loop fuzzy control system of segmented cooling is introduced, in order to solve the problem existing in the control of strip crown in the hot rolling of aluminum alloys. The simple measurement equipment and control system is used, instead of the expensive and complex strip crown control system. First the adaptive PSO-BP neural network is trained with the actual data of rolling, and the temperature presetting model is set by completed training neural network based on the target crown of aluminum strip. Then, the fuzzy control rules of the segmented cooling system are designed according to the operation experience of worker and the results of theoretical analysis. The goal of control strip crown is achieved ultimately. Through the verification of the two high reversible hot rolling mill, the deviation value of roll temperature can be controlled within ±4 ℃ and the proportion of aluminum strip crown within the scope of target crown(20~40 μm) is 95% on the aluminum strip with longitudinal. The method takes full advantage of the control ability of segmented cooling system for strip crown.
[1] 刘静安, 盛春磊, 朱英. 铝合金轧制设备国产化现状及发展趋势[J]. 轻合金加工技术, 2015, 43(1)：8-14.
LIU Jingan, SHENG Chunlei, ZHU Ying. Present situation and development trend of localization of aluminum alloy rolling equipments[J]. Light Alloy Fabrication Technology, 2015, 43(1)：8-14.
[2] 王祝堂. 中国铝平扎产品工业概览[J]. 有色金属加工, 2013, 42(6)：1-10.
WANG Zhutang. The overview of aluminum products in China[J]. Nonferrous Metals Processing, 2013, 42(6)：1-10.
[3] 陆国梁. 1+4热连轧生产线的自动凸度控制系统[J]. 铝加工, 2006, 171(5)：37-42.
LU Guoliang. Automatic camber control system of 1+4 hot tandem line[J]. Aluminum Fabrication, 2006, 171(5)：37-42.
[4] JIANG Mian, LI Xuejun, WU Jigang, et al. A precision on-line model for the prediction of thermal crown in hot rolling processes[J]. International Journal of Heat and Mass Transfer, 2014(78)：967-973.
[5] RIAHIFAR R, SERAJZADEH S. Three-dimensional model for hot rolling of aluminum alloys[J]. Materials and Design, 2007, 28(8)：2366-2372.
[6] BAGHERIPOOR M, BISADI H. Effects of rolling parameters on temperature distribution in the hot rolling of aluminum strips[J]. Applied Thermal Engineering, 2011, 31(10)：1556-1565.
[7] GUO Xinyu, HE Anrui, SHAO Jian, et al. Modeling and simulation of subsectional cooling system during hot aluminum rolling[J]. Journal of Mechanical Engineering, 2013, 49(4)：70-74.
[8] NAKAGAWA K. Heat crown of work rolls during aluminum hot rolling[J]. Sumitomo Light Metal Technical Reports, 1980, 21(1)：45-51.
[9] 郭新宇, 何安瑞, 邵健, 等. 铝热连轧分段冷却系统建模与仿真[J]. 机械工程学报, 2013, 49(4)：71-75.
GUO Xinyu, HE Anrui, SHAO Jian, et al. Modeling and simulation of subsectional cooling system during hot aluminum rolling[J]. Journal of Mechanical Engineering, 2013, 49(4)：71-75.
[10] 陈超超, 邵健, 何安瑞. 热轧带钢温度场在线计算方法研究[J]. 机械工程学报, 2014, 50(14)：135-142.
CHEN Chaochao, SHAO Jian, HE Anrui. Research on online calculation methods of temperature field of hot strip[J]. Journal of Mechanical Engineering, 2014, 50(14)：135-142.
[11] JIA Chunyu, SHAN Xiuying, LIU Hongmin, et al. Fuzzy neural model for flatness pattern recognition[J]. Journal of Iron and Steel Research, International, 2008, 15(6)：33-38.
[12] WANG Haiyong, HUANG Changqing, DENG Hua. Prediction on transverse thickness of hot rolling aluminum strip based on BP neural network[J]. Applied Mechanics and Materials, 2012, 157-158：78-83.
[13] SIKDAR S, KUMARI S. Neural network model of the profile of hot-rolled strip[J]. Advanced Manufacturing Technology, 2009,42(5)：450-462.
[14] ABBASPOUR M, SABOONCHI A. Work roll thermal expansion control in hot strip mill[J]. Applied Mathematical Modelling, 2008, 32(12)：2652-2699.
[15] JIA Chunyu, LIU Hongmin, ZHOU Huifeng. Fuzzy control system of hydraulic roll bending based on genetic neural network[J]. Journal Iron & Steel Research, International, 2005, 12(3)：22-27.
[16] 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 Xiuling, CHEN Lijie, PANG Zongpeng, 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.
[17] 刘建昌, 陈莹莹, 张瑞友. 基于PSO-BP网络的板形智能控制器[J]. 控制理论与应用, 2007, 24(4)：674-678.
LIU Jianchang, CHEN Yingying, ZHANG Ruiyou. Intelligent flatness-controller based on PSO-BP network[J]. Control Theory & Applications, 2007, 24(4)：674-678.
