锻压技术

基于稀疏自编码器与自组织映射网络的轧机颤振预警方法

英文标题：Rolling mill chatter warning method based on sparse auto-encoder and self-organizing map network

单位：1. 燕山大学 2. 南昌工学院

分类号：O332; TH113

出版年,卷(期):页码：2023,48(1):171-178

摘要：

颤振是轧机生产过程中常见的问题之一, 严重影响轧机的生产效率。为实现轧机颤振状态的实时监测, 预防轧机发生颤振, 提出了一种轧机颤振预警方法。该方法利用稀疏自编码器对轧机的振动数据进行降维融合, 并且通过自组织映射网络构建能够准确地反应轧机振动趋势的特征指标; 同时, 以轧机正常运行状态的数据为基准, 通过3σ 准则设定合理有效的阈值。实验结果表明: 所构造的轧机振动趋势特征指标以及设定的报警阈值能够及时发现轧机振动趋势的变化, 并在振动达到峰值之前进行报警。最后, 将提出的SAE-SOM 模型与AE-SOM 模型进行比较, 结果表明, SAE-SOM 模型更加稳定且能够更早发现振动状态的异常变化。

Chatter is one of the common problems in the production process of rolling mill, which seriously affects the production efficiency of rolling mill. Therefore, in order to monitor the chatter state of rolling mill in real-time and prevent the happening of chatter of rolling mill, a chatter warning method of rolling mill was proposed, which used the sparse auto-encoder to reduce the dimension of rolling mill vibration data and constructed the characteristic index that could accurately reflect the vibration trend of rolling mill through the self-organizing map network. At the same time, based on the data of normal running state for rolling mill, a reasonable and effective threshold value was set by 3σ criterion. The experimental results show that the constructed characteristic index of the vibration trend for rolling mill and the set alarm threshold value can detect the change of the vibration trend for rolling mill in time and give an alarm before the vibration reaches the peak value. Finally, compared with the AE-SOM model, the results show that the SAE-SOM model is more stable and can detect the abnormal changes in the vibration state earlier.

基金项目：

国家自然科学基金资助项目(61973262); 河北省自然科学基金资助项目(E2019203146); 中央引导地方科技发展资金项目(216Z2102G); 江西省教育厅科学技术研究项目(GJJ212504)

作者简介: 时培明(1979-), 男, 博士, 教授 E-mail: spm@ ysu. edu. cn

参考文献：

[1] 　林鹤, 邹家祥, 岳海龙. 四辊冷轧机第三倍频程颤振[J]. 钢铁, 1999, (12): 56-59.

Lin H, Zou J X, Yue H L. Chatter in the third frequency range of four-high cold mill [J]. Iron and Steel, 1999, (12): 56-59.

[2] 　邢德茂, 姚利辉, 李学通. 2030 mm 冷连轧机组板形预报及影响因素研究[J]. 塑性工程学报, 2021, 28 (3): 210-216.

Xing D M, Yao L H, Li X T. Research on shape prediction and influencing factors of 2030 mm tandem cold rolling mill [J]. Journal of Plastic Engineering, 2021, 28 (3): 210-216.

[3] 　侯福祥, 张杰, 曹建国, 等. 带钢冷轧机振动问题的研究进展及评述[J]. 钢铁研究学报, 2007, 19 (10): 6-10, 39.

Hou F X, Zhang J, Cao J G, et al. Research progress and review on vibration of cold strip mill [J]. Journal of Iron and Steel Research, 2007, 19 (10): 6-10, 39.

[4] 　Tlusty J, Chandra G, Critchley S, et al. Chatter in cold rolling [J]. CIRP Annals-Manufacturing Technology, 1982, 31 (1):195-199.

[5] 　钟掘, 唐华平. 高速轧机若干振动问题———复杂机电系统耦合动力学研究[J]. 振动、测试与诊断, 2002, 22 (1): 1-8.

Zhong J, Tang H P. Vibration problems of high speed rolling mill-Study on coupling dynamics of complex electromechanical system [J]. Journal of Vibration, Measurement & Fault, 2002, 22 (1):1-8.

[6] 　王长松, 陈志健, 陈先霖. 冷带轧机颤振现象的分析与仿真[J]. 北京科技大学学报, 1991, (1): 15-19.

Wang C S, Chen Z J, Chen X L. Analysis and simulation of chatter in cold strip mill [J]. Journal of University of Science and Technology Beijing, 1991, (1): 15-19.

[7] 　杨晋玲, 段牧忻. 轧机垂直振动特性研究及测试分析[J]. 锻压技术, 2021, 46 (7): 229-236.

Yang J L, Duan M X. Research and test analysis of vertical vibration characteristics of rolling mill [J]. Forging & Stamping Technology, 2021, 46 (7): 229-236.

[8] 　侯东晓, 陈浩, 刘彬, 等. 轧机辊系垂直非线性参激振动特性分析[J]. 振动与冲击, 2009, 28 (11): 1-5.

Hou D X, Chen H, Liu B, et al. Analysis of vertical nonlinear parametric vibration characteristics of rolling mill roll system [J]. Vibration and Impact, 2009, 28 (11): 1-5.

[9] 　王桥医, 崔明超, 王瀚, 等. 基于辊系多模态模式的连轧机机架间耦合振动系统模型的建立及仿真分析[J]. 中南大学学报: 自然科学版, 2020, 51 (10): 2834-2843.

Wang Q Y, Cui M C, Wang H, et al. Model establishment and simulation analysis of coupled vibration system between stands of continuous rolling mill based on roll system multi-mode mode [J]. Journal of Central South University: Science and Technology, 2020, 51 (10): 2834-2843.

[10] 彭艳. 冶金轧制设备技术数字化智能化发展综述[J]. 燕山大学学报, 2020, 44 (3): 218-237.

Peng Y. Review on digital and intelligent development of metallurgical rolling equipment technology [J]. Journal of Yanshan University, 2020, 44 (3): 218-237.

[11] 闫晓强. 热连轧机机电液耦合振动控制[J]. 机械工程学报,2011, 47 (17): 61-65.

Yan X Q. Electromechanical hydraulic coupling vibration control of hot strip rolling mill [ J]. Journal of Mechanical Engineering, 2011, 47 (17): 61-65.

[12] 王鑫鑫, 闫晓强. 基于扩张状态观测器的轧机振动抑振器研究[J]. 振动与冲击, 2019, 38 (5): 1-6.

Wang X X, Yan X Q. Research on rolling mill vibration suppressor based on extended state observer [ J]. Vibration and Shock, 2019, 38 (5): 1-6.

[13] 董志奎, 梁朋伟, 禚超越, 等. 基于DBN 算法的热轧高强钢薄板轧机振动预报研究[ J]. 矿冶工程, 2020, 40 (4):135-141, 144.

Dong Z K, Liang P W, Zhuo C Y, et al. Research on vibration prediction of hot rolled high strength steel sheet mill based on DBN algorithm [ J]. Mining and Metallurgy Engineering, 2020, 40(4): 135-141, 144.

[14] 万年红, 姚寿军, 全基哲, 等. 带钢振动纹测试和预警系统[J]. 钢铁研究学报, 2012, (S1): 36-39.

Wan N H, Yao S J, Quan J Z, et al. Strip steel vibration crack test and early warning system [J] Journal of Iron and Steel Research, 2012, (S1): 36-39.

[15] 米凯夫, 张杰, 曹建国, 等. 基于小波和小波分形的冷连轧机振动识别方法[J]. 北京科技大学学报, 2013, 35 (8):1064-1071.

Mi K F, Zhang J, Cao J G, et al. Vibration identification method of tandem cold rolling mill based on wavelet and wavelet fractal [J]. Journal of Beijing University of Science and Technology,

2013, 35 (8): 1064-1071.

[16] Lu X, Sun J, Song Z, et al. Prediction and analysis of cold rolling mill vibration based on a data-driven method [J]. Applied Soft Computing, 2020, 96: 106706.

[17] 胡昭华, 宋耀良. 基于Autoencoder 网络的数据降维和重构[J]. 电子与信息学报, 2009, 31 (5): 1189-1192.

Hu Z H, Song Y L. Data dimensionality reduction and reconstruction based on autoencoder network [J]. Journal of Electronics and Information, 2009, 31 (5): 1189-1192.

[18] Li Y, Ren J, Liu J, et al. Deep sparse autoencoder prediction model based on adversarial learning for cross-domain recommendations [J]. Knowledge-Based Systems, 2021, 220 (20): 106948.

[19] Kohonen T. The self-organizing map [ J]. IEEE Proc Icnn, 1990, 1 (1-3): 1-6.

[20] Pan Y, Hong R, Chen J, et al. Hybrid DBN-SOM-PF-based prognostic approach of remaining useful life for wind turbine gearbox [J]. Renewable Energy, 2020, 152 (6): 138-154.

[21] Hong S, Zhou Z, Zio E, et al. Condition assessment for the performance degradation of bearing based on a combinatorial feature extraction method [ J]. Digital Signal Processing, 2014, 27(1): 159-166.

[22] 李一青, 訾艳阳, 郎倩, 等. 基于多特征融合的轧机自激振动预警方法[J]. 振动. 测试与诊断, 2013, 33 (S1): 141-144, 226.

Li Y Q, Zi Y Y, Lang Q, et al. Early warning method of rolling mill self-excited vibration based on multi feature fusion [J]. Vibration. Test and Diagnosis, 2013, 33 ( S1 ): 141 - 144,

226. 　　　

[23] 高萌, 吴海锋, 沈勇, 等. 捣固车磁力信号峰值降噪整形检测方法研究[ J]. 传感技术学报, 2020, 33 ( 4): 546 -551. 　　　

Gao M, Wu H F, Shen Y, et al. Research on peak value noise reduction and shaping detection method of magnetic signal of tamping truck [J]. Journal of Sensing Technology, 2020, 33 (4): 546-551.

[24] Guo L, Li N, Jia F, et al. A recurrent neural network based health indicator for remaining useful life prediction of bearings [J]. Neurocomputing, 2017, 240 (C): 98-109.

服务与反馈：

【文章下载】【加入收藏】