Flatness Prediction of Cold Rolled Strip Based on Deep Neural Network with Improved Activation Function

被引:10
|
作者
Liu, Jingyi [1 ]
Song, Shuni [1 ]
Wang, Jiayi [1 ]
Balaiti, Maimutimin [2 ]
Song, Nina [2 ]
Li, Sen [2 ]
机构
[1] Northeastern Univ, Coll Sci, Shenyang 110819, Peoples R China
[2] Northeastern Univ, Coll Informat Sci & Engn, Shenyang 110819, Peoples R China
基金
中国国家自然科学基金;
关键词
cold rolled strip; deep neural network; Swish activation function; non-convex optimization; batch normalization;
D O I
10.3390/s22020656
中图分类号
O65 [分析化学];
学科分类号
070302 ; 081704 ;
摘要
With the improvement of industrial requirements for the quality of cold rolled strips, flatness has become one of the most important indicators for measuring the quality of cold rolled strips. In this paper, the strip production data of a 1250 mm tandem cold mill in a steel plant is modeled by an improved deep neural network (the improved DNN) to improve the accuracy of strip shape prediction. Firstly, the type of activation function is analyzed, and the monotonicity of the activation function is deemed independent of the convexity of the loss function in the deep network. Regardless of whether the activation function is monotonic, the loss function is not strictly convex. Secondly, the non-convex optimization of the loss functionextended from the deep linear network to the deep nonlinear network, is discussed, and the critical point of the deep nonlinear network is identified as the global minimum point. Finally, an improved Swish activation function based on batch normalization is proposed, and its performance is evaluated on the MNIST dataset. The experimental results show that the loss of an improved Swish function is lower than that of other activation functions. The prediction accuracy of a deep neural network (DNN) with an improved Swish function is 0.38% more than that of a deep neural network (DNN) with a regular Swish function. For the DNN with the improved Swish function, the mean square error of the prediction for the flatness of cold rolled strip is reduced to 65% of the regular DNN. The accuracy of the improved DNN is up to and higher than the industrial requirements. The shape prediction of the improved DNN will assist and guide the industrial production process, reducing the scrap yield and industrial cost.
引用
收藏
页数:16
相关论文
共 50 条
  • [1] Flatness Prediction of Cold Rolled Strip Based on EM-TELM
    Liu, Jingyi
    Wan, Lushan
    Xiao, Dong
    [J]. IEEE ACCESS, 2021, 9 : 51484 - 51493
  • [2] Prediction of head thickness of hot-rolled strip based on deep neural network
    Yuan, Shangbin
    Sun, Jie
    Di, Hongshuang
    [J]. 2022 34TH CHINESE CONTROL AND DECISION CONFERENCE, CCDC, 2022, : 646 - 653
  • [3] Flatness control of cold rolled strip based on relay optimisation
    Wang, P. F.
    Zhang, Z. J.
    Sun, J.
    Zhang, D. H.
    Liu, H. M.
    Gao, X. L.
    [J]. IRONMAKING & STEELMAKING, 2018, 45 (02) : 166 - 175
  • [4] Prediction of tandem cold-rolled strip flatness based on Attention-LSTM model
    Chen, Yafei
    Peng, Lianggui
    Wang, Yu
    Zhou, Yilin
    Li, Changsheng
    [J]. JOURNAL OF MANUFACTURING PROCESSES, 2023, 91 : 110 - 121
  • [5] Prediction of coiling temperature of hot rolled strip based on neural network
    Wang, YQ
    Wang, HF
    Zhu, DD
    [J]. Proceedings of the Sixth International Conference on Fluid Power Transmission and Control, 2005, : 731 - 735
  • [6] FLATNESS DEFECTS IN COLD-ROLLED STRIP AND TESTING OF A FLATNESS MEASURING DEVICE
    BERGER, B
    THIES, H
    NEUSCHUTZ, E
    [J]. STAHL UND EISEN, 1975, 95 (22): : 1051 - 1056
  • [7] Prediction of mechanical properties of cold rolled strip based on improved extreme random tree
    Zhao, Yun-bao
    Song, Yong
    Li, Fei-fei
    Yan, Xian-le
    [J]. JOURNAL OF IRON AND STEEL RESEARCH INTERNATIONAL, 2023, 30 (02) : 293 - 304
  • [8] Prediction of mechanical properties of cold rolled strip based on improved extreme random tree
    Yun-bao Zhao
    Yong Song
    Fei-fei Li
    Xian-le Yan
    [J]. Journal of Iron and Steel Research International, 2023, 30 : 293 - 304
  • [9] Strip flatness prediction of cold rolling based on ensemble methods
    Yang, Wu-quan
    Zhao, Zhi-ting
    Zhu, Liang-yu
    Gao, Xun-yang
    Wang, Li
    [J]. JOURNAL OF IRON AND STEEL RESEARCH INTERNATIONAL, 2024, 31 (01) : 237 - 251
  • [10] Strip flatness prediction of cold rolling based on ensemble methods
    Wu-quan Yang
    Zhi-ting Zhao
    Liang-yu Zhu
    Xun-yang Gao
    Li Wang
    [J]. Journal of Iron and Steel Research International, 2024, 31 (1) : 237 - 251