-
摘要: 熱軋支持輥的健康狀態在帶鋼板形質量和軋制穩定性控制中起著關鍵作用，非線性、強耦合、少樣本等特點使得熱軋支持輥健康狀態的預測復雜，目前各大鋼廠仍以定期維護和事后維修為主。本文提出了一種支持輥虛擬健康指數的構建方法以及基于Copula函數的復雜工況健康狀態預測模型。首先結合支持輥彎竄輥數據表征支持輥健康狀態，再使用K-means聚類方法對支持輥工況進行劃分，將各工況下過程數據分別構建Copula預測模型，最后根據實際軋制計劃的排布順序融合各工況模型的預測結果。提出的基于Copula函數的預測模型在某鋼廠1780熱連軋產線得到應用，結果表明，該模型能夠準確有效的按照軋制計劃實現支持輥的健康狀態預測，以更科學的策略指導支持輥更換維護。Abstract: The health condition of hot-rolling back-up rolls plays a key role in controlling the strip profile quality and rolling stability. The characteristics of nonlinearity, strong coupling, and the use of limited samples complicate the prediction of the back-up roll health state. The current back-up roll replacement strategy of each steel mill is generally determined according to a certain rolling time or rolling kilometer, and such a maintenance mode is based on experience. In actual experience, due to different strip specifications in each rolling cycle, the degrees of wear on the back-up rolls are different. Regular maintenance methods may easily lead to excessive wear of the back-up rolls and reduce the quality of the strip shape at the end of the unit, or premature roll replacement wastes the back-up roll performance. This paper proposed a construction method for the back-up roll virtual health index and a Copula function–based model for predicting the health condition of complex working conditions. The health condition of a pair of back-up rolls was characterized by combining roll bending and shifting data, and the back-up roll condition was divided by the K-means clustering method. The Copula prediction model was constructed using the process data under each working condition, and finally, according to the actual rolling schedule, the arrangement order combines the prediction results of the working conditions. The production performance data of a 1780-mm hot rolling line were used to verify the results. The results show that the proposed Copula-based prediction model can accurately and effectively predict the health condition of the back-up roll according to the rolling schedule; thus, it can serve as the basis of a more scientific strategy to guide the replacement and maintenance of the back-up roll.
-
Key words:
- back-up roll /
- health prognostics /
- Copula function /
- data-driven /
- profile and flatness
-
圖 1 F7機架VHI數據（1#）表現出隨軋制計劃推進而上升的趨勢
Figure 1. Rising trend of F7 stand VHI data (1#) with the rolling schedule
圖 4 工況1支持輥VHI數據擬合降噪后結果示意圖
Figure 4. Result of noise reduction after fitting VHI data of working condition 1
圖 5 使用Copula函數預測支持輥健康狀態的建模流程圖
Figure 5. Flow chart for predicting the health of back-up roll using Copula function
圖 6 對單工況訓練集數據劃分退化等級示意圖
Figure 6. Data degradation level for single-working-condition training set
圖 7 處于不同退化等級下的分布模型適用于不同類型的Copula函數描述。（a）T12-T50, Gumbel; （b） T26?T50, Frank; （c） T46?T50, Clayton
Figure 7. Distribution models at different degradation levels fit for different types of Copula function descriptions: (a) T12?T50, Gumbel; (b) T26?T50, Frank; (c) T46?T50, Clayton
圖 8 提高模型適應性的平移處理。（a）過早失效時Copula模型無法預測；（b）Copula模型平移
Figure 8. Translation processing to improve model adaptability: (a) Copula model is unpredictable at premature failure; (b) translation of copula model
圖 10 經過VHI變尺度處理的單工況預測結果
Figure 10. Single-condition VHI prediction result after scale conversion
表 1 某鋼廠1780熱連軋產線F7機架支持輥使用情況統計
Table 1. Statistics on the use of F7 back-up roll in a 1780 hot rolling line
Data number Total number of rolled strips Total rolling weight/t Total rolling length /km 1# 15016 336000 12015 2# 16024 356000 13216 3# 17654 388900 15015 4# 14168 308000 12282 5# 16291 362300 13596 
下載: 導出CSV
中文字幕在线观看表 2 復雜工況下Copula模型融合預測結果
Table 2. Copula model fusion prediction results under complex conditions
Test set number Actual number of rolled strips Actual VHI Predict VHI Model error/% 5# 16291 0.726 0.769 5.85 4# 14168 0.831 0.758 ?8.74 3# 17654 0.909 0.976 7.34 2# 16024 0.825 0.862 4.42 1# 15016 0.891 0.815 ?8.54
下載: 導出CSV
-
參考文獻
[1] Shen Y M. Introduction of application technology of backup roll in Baosteel. Steel Roll, 2004, 21(1): 55 doi: 10.3969/j.issn.1003-9996.2004.01.018沈一鳴. 寶鋼支撐輥的使用技術. 軋鋼, 2004, 21(1):55 doi: 10.3969/j.issn.1003-9996.2004.01.018 [2] He A R, Zhang Q D, Cao J G, et al. Effect of back-up roll profile in hot wide strip mill on the strip profile and flatness. J Univ Sci Technol Beijing, 1999, 21(6): 565 doi: 10.3321/j.issn:1001-053X.1999.06.015何安瑞, 張清東, 曹建國, 等. 寬帶鋼熱軋支持輥輥形變化對板形的影響. 北京科技大學學報, 1999, 21(6):565 doi: 10.3321/j.issn:1001-053X.1999.06.015 [3] Li Z W, Gao S H, Han L W, et al. Chenggang Company 1780 save backup roll roll change time // Proceedings of the 10th CSM Steel Congress & the 6th Baosteel Biennial Academic Conference. Shanghai, 2015: 623李子文, 高少華, 韓立偉, 等. 承鋼1780節約支撐輥換輥時間//第十屆中國鋼鐵年會暨第六屆寶鋼學術年會論文集. 上海, 2015: 623 [4] Shi Z Y, Jing T, Dong B Q, et al. Application optimization of BUR according to the mill features // 2011 CSM Annual Meeting Proceedings. Beijing, 2011: 2882史志勇, 荊濤, 董寶權, 等. 結合熱軋機組特點合理使用支撐輥//第八屆中國鋼鐵年會論文集. 北京, 2011: 2882 [5] Alford L D. The problem with aviation COTS. IEEE Aerosp Electron Syst Mag, 2001, 16(2): 33 doi: 10.1109/62.904242 [6] Gartner D L, Dibbert S E. Application of integrated diagnostic process to non-avionics systems // 2001 IEEE Autotestcon Proceedings. IEEE Systems Readiness Technology Conference. Valley Forge, 2001: 229 [7] Peng Y, Liu D T, Peng X Y. A review: prognostics and health management. J Electron Meas Instrum, 2010, 24(1): 1 doi: 10.3724/SP.J.1187.2010.00001彭宇, 劉大同, 彭喜元. 故障預測與健康管理技術綜述. 電子測量與儀器學報, 2010, 24(1):1 doi: 10.3724/SP.J.1187.2010.00001 [8] Ma S S, Xu A H, Zhang Q X, et al. The security monitoring and health management system for UAV. Aviat Mainten Eng, 2009(4): 63 doi: 10.3969/j.issn.1672-0989.2009.04.027馬颯颯, 許愛華, 張群興, 等. 無人機安全監控與健康管理系統研究. 航空維修與工程, 2009(4):63 doi: 10.3969/j.issn.1672-0989.2009.04.027 [9] Yang L F, Wang L, Feng J C. Maintenance support of missile based on PHM technology. J Naval Aeron Astron Univ, 2010, 25(4): 447 doi: 10.3969/j.issn.1673-1522.2010.04.020楊立峰, 王亮, 馮佳晨. 基于PHM技術的導彈維修保障. 海軍航空工程學院學報, 2010, 25(4):447 doi: 10.3969/j.issn.1673-1522.2010.04.020 [10] Zhang D, Feng Z P. Fault diagnosis of rolling bearings based on variational mode decomposition and calculus enhanced energy operator. Chin J Eng, 2016, 38(9): 1327張東, 馮志鵬. 基于變分模式分解和微積分增強能量算子的滾動軸承故障診斷. 工程科學學報, 2016, 38(9):1327 [11] Liu Y B, Zhou Y K, Feng Z P. Application of morphological component analysis for rolling element bearing fault diagnosis. Chin J Eng, 2017, 39(6): 909劉永兵, 周亞凱, 馮志鵬. 形態分量分析在滾動軸承故障診斷中的應用. 工程科學學報, 2017, 39(6):909 [12] Zhang J L, Yang J H, Tang C Q, et al. Bearing fault diagnosis by stochastic resonance method in periodical potential system. Chin J Eng, 2018, 40(8): 989張景玲, 楊建華, 唐超權, 等. 基于周期勢系統隨機共振的軸承故障診斷. 工程科學學報, 2018, 40(8):989 [13] Zhao C, Feng Z P. Localized fault identification of planetary gearboxes based on multiple-domain manifold. Chin J Eng, 2017, 39(5): 769趙川, 馮志鵬. 基于多域流形的行星齒輪箱局部故障識別. 工程科學學報, 2017, 39(5):769 [14] Zhang D, Feng Z P. Application of iterative generalized short-time Fourier transform to fault diagnosis of planetary gearboxes. Chin J Eng, 2017, 39(4): 604張東, 馮志鵬. 迭代廣義短時Fourier變換在行星齒輪箱故障診斷中的應用. 工程科學學報, 2017, 39(4):604 [15] Gugulothu N, Tv V, Malhotra P, et al. Predicting remaining useful life using time series embeddings based on recurrent neural networks // ACM SIGKDD Workshop on Machine Learning for Prognostics and Health Management. Halifax, 2017 [16] Guo L, Li N P, Jia F, et al. A recurrent neural network based health indicator for remaining useful life prediction of bearings. Neurocomputing, 2017, 240: 98 doi: 10.1016/j.neucom.2017.02.045 [17] Huang H Z, Wang H K, Li Y F, et al. Support vector machine based estimation of remaining useful life: current research status and future trends. J Mech Sci Technol, 2015, 29(1): 151 doi: 10.1007/s12206-014-1222-z [18] Liu J, Zio E. SVM hyperparameters tuning for recursive multi-step-ahead prediction. Neural Comput Appl, 2017, 38: 3749 [19] Gao M Z, Xu A Q, Xu Q. Online condition prediction of electronic equipment based on relevance vector machine with adaptive Kernel learning. J Ordnance Equip Eng, 2017, 38(11): 108 doi: 10.11809/scbgxb2017.11.024高明哲, 許愛強, 許晴. 基于aRVM的電子設備狀態在線預測方法. 兵器裝備工程學報, 2017, 38(11):108 doi: 10.11809/scbgxb2017.11.024 [20] Zermani S, Dezan C, Chenini H, et al. FPGA implementation of Bayesian network inference for an embedded diagnosis // 2015 IEEE Conference on Prognostics and Health Management. Austin, 2015: 1 [21] Zermani S, Dezan C, Euler R, et al. Bayesian network-based framework for the design of reconfigurable health management monitors // 2015 NASA/ESA Conference on Adaptive Hardware and Systems. Montreal, 2015: 1 [22] Wang T Y, Yu J B, Siegel D, et al. A similarity-based prognostics approach for remaining useful life estimation of engineered systems // International Conference on Prognostics and Health Management. Denver, 2008: 1 [23] Zhu Y P. Generalization and application of NDT technology in roller testing. Mech Electr Eng Technol, 2016, 45(5): 121 doi: 10.3969/j.issn.1009-9492.2016.05.034朱揚普. 無損檢測技術在軋輥檢測中的推廣及應用. 機電工程技術, 2016, 45(5):121 doi: 10.3969/j.issn.1009-9492.2016.05.034 [24] He A R, Shao J, Sun W Q. Theory and Practice of Shape Control. Beijing: Metallurgical Industry Press, 2016何安瑞, 邵鍵, 孫文權. 板形控制理論與實踐. 北京: 冶金工業出版社, 2016 [25] Nelsen R B. An Introduction to Copulas. Technometrics, 2000, 42(3): 317 -
點擊查看大圖
計量
- 文章訪問數: 2696
- HTML全文瀏覽量: 1273
- PDF下載量: 78
- 被引次數: 0
