Abstract：The cold rolling force prediction results directly affect the rolling precision and product quality of the plate (belt). Because of the complicated cold rolling process and strong coupling of parameters, the model is not easy to be established and the actual deviation is large. An improved on-line prediction method of online sequential extreme learning machine is proposed. Using quantum behaved particle swarm optimization algorithm to optimize the weights and thresholds, according to the contribution of the hidden layer to the network output in the current training data, the topology structure of the network is adjusted and realized the self-organization of the structure and parameters. The experimental results show that the self-organizing online sequential extreme learning machine has a higher improvement in the training speed and precision than the artificial bee colony optimization in the reverse propagation neural network and incremental extreme learning machine based on the enhanced random search.
［1］Orowan E. The Calculation of Roll Pressure in Hot and Cold Flat Rolling［J］. Archive Proceedings of the Institution of Mechanical Engineers, 1943, 150:140-167.
［2］Kim D J, Kim Y C, Kim B M. Optimization of the irregular shape rolling process with an artificial neural network［J］. Journal of Materials Processing Technology, 2001, 113(1-3):131-135.
［3］Reddy N V, Suryanarayana G. A set-up model for tandem cold rolling mills［J］. Journal of Materials Processing Technology, 2001, 116(2): 269-277.
［4］赵志伟, 杨景明, 车海军, 等. 基于人工蜂群算法与反向传播神经网络的铝热连轧轧制力预测［J］. 计量学报, 2014, 35(2): 157-160.
Zhao Z W, Yang J M, Che H J, et al. Prediction of Rolling Force Based on Artificial Bee Colony Algorithm and Back Propagation Neural Network in Aluminum Hot Tandem Rolling［J］. Acta Metrologica Sinica, 2014, 35(2): 157-160.
［5］Zhao Z W, Yang J M, Che H, et al. Application of artificial bee colony algorithm to select architecture of a optimal neural network for the prediction of rolling force in hot strip rolling process［J］. Journal of Chemical and Pharmaceutical Research, 2013, 5(9): 563-570.
［6］Huang G B, Chen L. Enhanced random search based incremental extreme learning machine［J］.Neurocomputing, 2008, 71(16-18): 3460-3468.
［7］Jun Y, Er M J. An Enhanced Online Sequential Extreme Learning Machine Algorithm［C］//IEEE.2008 Chinese Control And Decision Conference. Yantai,China, 2008: 2 902-2 907.
［8］王荣军, 马立峰, 蒋亚平, 等. AZ31B镁合金变形抗力数学模型的建立［J］. 材料热处理学报, 2017, 38(2): 168-172.
Wang R J, Ma L F, Jiang Y P, et al. Mathematical model of deformation resistance for AZ31B magnesium alloy［J］. Transactions of Materials and Heat Treatment, 2017, 38(2): 168-172.
［9］Huang G B, Zhu Q Y, Siew C K. Extreme learning machine: Theory and applications［J］. Neurocomputing, 2006, 70(1-3): 489-501.
［10］何群, 王红, 江国乾, 等. 基于相关主成分分析和极限学习机的风电机组主轴承状态监测研究［J］. 计量学报, 2018, 39(1): 89-93.
He Q, Wang H, Jiang G Q, et al. Research of Wind Turbine Main Bearing Condition Monitoring Based on Correlation PCA and ELM［J］. Acta Metrologica Sinica, 2018, 39(1): 89-93.
［11］杨景明, 马凤艳, 车海军, 等. 基于最小二乘支持向量机的反向建模的铝合金变形抗力模型［J］. 计量学报, 2013, 34(6): 532-536.
Yang J M, Ma F Y, Che H J, et al. Deformation Resistance Model of Aluminum Alloy Based on Least Squares Support Vector Machine Reversed Modeling Method［J］. Acta Metrologica Sinica, 2013, 34(6): 532-536.
［12］孙浩, 杨景明, 呼子宇, 等. 基于改进鱼群算法的铝热连轧摩擦系数模型研究［J］. 计量学报, 2016, 37(1): 53-55.
Sun H, Yang J M, Hu Z Y, et al. Study on Aluminum Hot Rolling Model of Friction Coefficient Based on Improved Fish Swarm Algorithm［J］. Acta Metrologica Sinica, 2016, 37(1): 53-55.
［13］Sun J, Wu X, Palade V, et al. Convergence analysis and improvements of quantum-behaved particle swarm optimization［J］. Information Sciences, 2012, 193: 81-103.
［14］Shao Z, Er M J. An online sequential learning algorithm for regularized Extreme Learning Machine［J］. Neurocomputing, 2016, 173: 778-788.
［15］Nwachukwu P U, Oluwole O O. Effects of rolling process parameters on the mechanical properties of hot-rolled St60Mn steel［J］. Case Studies in Construction Materials, 2017, 6: 134-146.
［16］李荣雨, 戚桂洪. 基于CSSE-OSELM算法的软测量建模及其工业应用［J］. 计量学报, 2017, 38(5): 650-655.
Li R Y,Qi G H. Soft Sensor Modeling Based on CSSE-OSELM Algorithm and Its Industrial Application［J］. Acta Metrologica Sinica, 2017, 38(5): 650-655.
［17］韩红桂, 乔俊飞, 薄迎春. 基于信息强度的RBF神经网络结构设计研究［J］. 自动化学报, 2012, 38(7): 1083-1090.
Han H G, Qiao J F, Bo Y C. On Structure Design for RBF Neural Network Based on Information Strength［J］, Acta Automatica Sinica, 2012, 38(7): 1083-1090.
2019, Vol. 40 
