A fault detection method based on stacking the SAE-SRBM for nonstationary and stationary hybrid processes

• Autorzy: Huang Lei , Ren Hao , Chai Yi , Qu Jianfeng

Identyfikatory

DOI

10.34768/amcs-2021-0003

• Języki publikacji: EN

Abstrakty

EN

This paper proposes a fault detection method by extracting nonlinear features for nonstationary and stationary hybrid industrial processes. The method is mainly built on the basis of a sparse auto-encoder and a sparse restricted Boltzmann machine (SAE-SRBM), so as to take advantages of their adaptive extraction and fusion on strong nonlinear symptoms. In the present work, SAEs are employed to reconstruct inputs and accomplish feature extraction by unsupervised mode, and their outputs present a knotty problem of an unknown probability distribution. In order to solve it, SRBMs are naturally used to fuse these unknown probability distribution features by transforming them into energy characteristics. The contribution of this method is the capability of further mining and learning of nonlinear features without considering the nonstationary problem. Also, this paper introduces a method of constructing labeled and unlabeled training samples while maintaining time series features. Unlabeled samples can be adopted to train the part for feature extraction and fusion, while labeled samples can be used to train the classification part. Finally, a simulation on the Tennessee Eastman process is carried out to demonstrate the effectiveness and excellent performance on fault detection for nonstationary and stationary hybrid industrial processes.

Słowa kluczowe

EN

fault detection; sparse autoencoder; sparse restricted Boltzmann machine; hybrid industrial processes

PL

detekcja błędów; autoenkoder; proces przemysłowy

• Wydawca: Oficyna Wydawnicza Uniwersytetu Zielonogórskiego
• Czasopismo: International Journal of Applied Mathematics and Computer Science
• Rocznik: 2021
• Tom: Vol. 31, no. 1
• Strony: 29--43
• Opis fizyczny: Bibliogr. 49 poz., rys., tab.

Twórcy

autor

Huang Lei

• School of Computer Science and Technology, Huaiyin Normal University, Huaian City, Jiangsu, 223300, China

autor

Ren Hao

• School of Automation, Chongqing University, Chongqing City, 400044, China; Peng Cheng Laboratory, Shenzhen City, Guangdong, 518000, China

autor

Chai Yi

• School of Automation, Chongqing University, Chongqing City, 400044, China; Key Laboratory of Complex System Safety and Control, Ministry of Education, Chongqing City, 400044, China

autor

Qu Jianfeng

• School of Automation, Chongqing University, Chongqing City, 400044, China; Key Laboratory of Complex System Safety and Control, Ministry of Education, Chongqing City, 400044, China

Bibliografia

• [1] Adil, M., Abid, M., Khan, A.Q., Mustafa, G. and Ahmed, N. (2016). Exponential discriminant analysis for fault diagnosis, Neurocomputing 171: 1344–1353, DOI: 10.1016/j.neucom.2015.07.099.
• [2] Byrski, W., Drapała, M. and Byrski, J. (2019). An adaptive identification method based on the modulating functions technique and exact state observers for modeling and simulation of a nonlinear MISO glass melting process, International Journal of Applied Mathematics and Computer Science 29(4): 739–757, DOI: 10.2478/amcs-2019-0055.
• [3] Chen, Q., Kruger, U. and Leung, A.Y.T. (2009). Cointegration testing method for monitoring nonstationary processes, Industrial and Engineering Chemistry Research 48(7): 3533–3543, DOI: 10.1021/ie801611s.
• [4] Chen, Z.W., Liu, C., Ding, S., Peng, T., Yang, C.H., Gui, W.H. and Shardt, Y. (2020a). A just-in-time-learning aided canonical correlation analysis method for multimode process monitoring and fault detection, IEEE Transactions on Industrial Electronics 68(6): 5259–5270, DOI: 10.1109/TIE.2020.2989708.
• [5] Chen, Z.W., Yang, C.H., Peng, T., Dan, H., Li, C. and Gui, W.H. (2020b). A cumulative canonical correlation analysis-based sensor precision degradation detection method, IEEE Transactions on Industrial Electronics 66(8): 6321–6330, DOI: 10.1109/TIE.2018.2873100.
• [6] Dickey, D.P. (1981). Likelihood ratio statistics for autoregressive time serries with a unit root, Econometrica 49(4): 1057–1072, DOI: 10.2307/1912517.
• [7] Dong, J., Zhang, K., Huang, Y., Li, G. and Peng, K.X. (2015). Adaptive total PLS based quality-relevant process monitoring with application to the Tennessee Eastman process, Neurocomputing 154: 77–85, DOI: 10.1016/j.neucom.2014.12.017.
• [8] Engle, R.F. and Granger, C.W.J. (1987). Co-integration and error correction: Representation, estimation, and testing, Econometrica 55(2): 251–276, DOI: 10.2307/1913236.
• [9] Firouzi, A., Yang, W. and Li, C.Q. (2018). Efficient solution for calculation of upcrossing rate of nonstationary Gaussian process, Journal of Engineering Mechanics 144(4): 04018015, DOI: 10.1061/(ASCE)EM.1943-7889.0001420.
• [10] Geng, Z.Q., Wang, Z., Hu, H.X., Han, Y.M., Lin, X.Y. and Zhong, Y.H. (2018). A fault detection method based on horizontal visibility graph-integrated complex networks: application to complex chemical processes, The Canadian Journal of Chemical Engineering 97(5): 1129–1138, DOI: 10.1002/cjce.23319.
• [11] He, Z.M., Zhou, H.Y., Wang, J.Q., Chen, Z.W., Wang, D.Y. and Yan, X. (2015). An improved detection statistic for monitoring the nonstationary and nonlinear processes, Chemometrics and Intelligent Laboratory Systems 145: 114–124, DOI: 10.1016/j.chemolab.2015.04.016.
• [12] Jiang, P., Hu, Z.X., Liu, J., Yu, S.N. and Wu, F. (2016). Fault diagnosis based on chemical sensor data with an active deep neural network, Sensors 16(10): 01695, DOI: 10.3390/s16101695.
• [13] Lee, H., Ekanadham, C. and Ng, A.Y. (2008). Sparse deep belief net model for visual area v2, Proceedings of the 21st International Conference on Neural Networks, NIPS’07, Vancouver, Canada, pp. 873–880.
• [14] Leng, J.W. and Jiang, P.Y. (2016). A deep learning approach for relationship extraction from interaction context in social manufacturing paradigm, Knowledge Based Systems 100: 188–199, DOI: 10.1016/j.knosys.2016.03.008.
• [15] Li, G., Qin, S.J. and Yuan, T. (2014). Nonstationarity and cointegration tests for fault detection of dynamic processes, IFAC Proceedings Volumes 47(3): 10616–10621, DOI: 10.3182/20140824-6-ZA-1003.00754.
• [16] Lin, Y.L., Kruger, U. and Chen, Q. (2017). Monitoring nonstationary dynamic systems using cointegration and common-trends analysis, Industrial and Engineering Chemistry Research 56(31): 8895–8905, DOI: 10.1021/acs.iecr.7b00011.
• [17] Lin, Y.L., Kruger, U., Gu, F.S., Ball, A. and Chen, Q. (2019a). Monitoring nonstationary and dynamic trends for practical process fault diagnosis, Control Engineering Practice 84: 139–158, DOI: 10.1016/j.conengprac.2018.11.020.
• [18] Lin, Y.L., Kruger, U., Gu, F.S., Balland, A. and Chen, Q. (2019b). Monitoring nonstationary processes using stationary subspace analysis and fractional integration order estimation, Industrial and Engineering Chemistry Research 58(16): 6486–6504, DOI: 10.1021/acs.iecr.8b05099.
• [19] Liu, Q. and Qin, S.J. (2016). Perspectives on big data modeling of process industries, Acta Automatica Sinica 42(2): 161–171.
• [20] Lu, W., Liang, B., Cheng, Y., Meng, D. and Zhang, T. (2017). Deep model based domain adaptation for fault diagnosis, IEEE Transactions on Industrial Electronics 64(99): 2296–2305, DOI: 10.1109/TIE.2016.2627020.
• [21] Ma, J., Li, G. and Zhou, D. (2018). Fault prognosis technology for non-Gaussian and nonlinear processes based on KICA reconstruction, Canadian Journal of Chemical Engineering 96(2): 515–520, DOI: 10.1002/cjce.23051.
• [22] Montmain, J., Labreuche, C., Imoussaten, A. and Trousset, F. (2015). Multi-criteria improvement of complex systems, Information Sciences 291: 61–84, DOI: 10.1016/j.ins.2014.08.027.
• [23] Orlov, A.A., Ushakov, A.A., Sovach, V.P. and Mymrina, D.F. (2017). Modeling of nonstationary processes during separation of multicomponent isotope mixtures, Separation Science and Technology 53(5): 796–806, DOI: 10.1080/01496395.2017.1399910.
• [24] Pröll, S., Lunze, J. and Jarmolowitz, F. (2018). From structural analysis to observer-based residual generation for fault detection, International Journal of Applied Mathematics and Computer Science 28(2): 233–245, DOI: 10.2478/amcs-2018-0017.
• [25] Ren, H., Chai, Y., Qu, J.F., Ye, X. and Tang, Q. (2018). A novel adaptive fault detection methodology for complex system using deep belief networks and multiple models: A case study on cryogenic propellant loading system, Neurocomputing 275: 2111–2125, DOI: 10.1016/j.neucom.2017.10.063.
• [26] Ren, H., Qu, J.F., Chai, Y., Tang, Q. and Ye, X. (2017). Deep learning for fault diagnosis: The state of the art and challenge, Control and Decision 32(8): 1343–1358, DOI: 10.13195/j.kzyjc.2016.1625.
• [27] Rumelhart, D.E., Hinton, G.E. and Williams, R.J. (1986). Learning representations by back-propagating errors, Nature 323(9): 533–536.
• [28] Sadough Vanini, Z.N., Khorasani, K. and Meskin, N. (2014). Fault detection and isolation of a dual spool gas turbine engine using dynamic neural networks and multiple model approach, Information Sciences 259(3): 234–251, DOI: 10.1016/j.ins.2013.05.032.
• [29] Severson, K., Chaiwatanodom, P. and Braatz, R.D. (2016). Perspectives on process monitoring of industrial systems, Annual Reviews in Control 42: 190–200, DOI: 10.1016/j.ifacol.2015.09.646.
• [30] Shang, J., Chen, M., Ji, H., Zhou, D., Zhang, H. and Li, M. (2017). Dominant trend based logistic regression for fault diagnosis in nonstationary processes, Control Engineering Practice 66: 156–168, DOI: 10.1016/j.conengprac.2017.06.011.
• [31] Souza, B.D., Kuhn, E.V. and Seara, R. (2019). A time-varying autoregressive model for characterizing nonstationary processes, IEEE Signal Processing Letters 26(1): 134–138, DOI: 10.1109/LSP.2018.2880086.
• [32] Sun, H., Zhang, S.M., Zhao, C.H. and Gao, F.R. (2017). A sparse reconstruction strategy for online fault diagnosis in nonstationary processes with no a priori fault information, Industrial and Engineering Chemistry Research 56(24): 6993–7008, DOI: 10.1021/acs.iecr.7b00156.
• [33] Sun, W.J., Shao, S.Y., Zhao, R., Yan, R.Q., Zhang, X.W. and Chen, X.F. (2016). A sparse auto-encoder-based deep neural network approach for induction motor faults classification, Measurement 89: 171–178, DOI: 10.1016/j.measurement.2016.04.007.
• [34] Utkin, L.V., Zaborovskii, V.S. and Popov, S.G. (2016). Detection of anomalous behavior in a robot system based on deep learning elements, Automatic Control and Computer Sciences 50(8): 726–733, DOI: 10.3103/S0146411616080319.
• [35] Wang, L.K., Zhao, X.Y., Pei, J.N. and Tang, G.Y. (2016). Transformer fault diagnosis using continuous sparse autoencoder, SpringerPlus 5, Article no.: 448, DOI: 10.1186/s40064-016-2107-7.
• [36] Worden, K., Baldacchino, T., Rowson, J. and Cross, E.J. (2016). Some recent developments in SHM based on nonstationary time series analysis, Proceedings of the IEEE 104(8): 1589–1603, DOI: 10.1109/JPROC.2016.2573596.
• [37] Xi, X.P., Chen, M.Y., Zhang, H.W. and Zhou, D.H. (2018). An improved non-Markovian degradation model with long-term dependency and item-to-item uncertainty, Mechanical Systems and Signal Processing 105: 467–480, DOI: 10.1016/j.ymssp.2017.12.017.
• [38] Xiong, Y. and Zuo, R. (2016). Recognition of geochemical anomalies using a deep autoencoder network, Computers and Geosciences 86: 75–82, DOI: 10.1016/j.cageo.2015.10.006.
• [39] Yan, W.W., Guo, P.J., Gong, L.A. and Li, Z.K. (2016). Nonlinear and robust statistical process monitoring based on variant autoencoders, Chemometrics and Intelligent Laboratory Systems 158: 31–40, DOI: 10.1016/j.chemolab.2016.08.007.
• [40] Yin, S., Ding, S.X., Haghani, A., Hao, H. and Zhang, P. (2012). A comparison study of basic data-driven fault diagnosis and process monitoring methods on the benchmark Tennessee Eastman process, Journal of Process Control 22(9): 1567–1581, DOI: 10.1016/j.jprocont.2012.06.009.
• [41] Zhang, Q., Yang, L.T. and Chen, Z.W. (2016). Deep computation model for unsupervised feature learning on big data, IEEE Transactions on Services Computing 9(1): 161–171, DOI: 10.1109/TSC.2015.2497705.
• [42] Zhang, Z.H., Jiang, T., Zhan, C.J. and Yang, Y.P. (2019). Gaussian feature learning based on variational autoencoder for improving nonlinear process monitoring, Journal of Process Control 75: 136–155, DOI: 10.1016/j.jprocont.2019.01.008.
• [43] Zhang, Z., Jiang, T., Li, S. and Yang, Y. (2018). Automated feature learning for nonlinear process monitoring-An approach using stacked denoising autoencoder and k-nearest neighbor rule, Journal of Process Control 64: 49–61, DOI: 10.1016/j.jprocont.2018.02.004.
• [44] Zhang, Z. and Zhao, J. (2017). A deep belief network based fault diagnosis model for complex chemical processes, Computers and Chemical Engineering 107: 395–407, DOI: 10.1016/j.compchemeng.2017.02.041.
• [45] Zhao, C.H. and Huang, B. (2018a). A full-condition monitoring method for nonstationary dynamic chemical processes with cointegration and slow feature analysis, American Institute of Chemical Engineers 64(5): 1662–1681, DOI: 10.1002/aic.16048.
• [46] Zhao, C.H. and Huang, B. (2018b). Incipient fault detection for complex industrial processes with stationary and nonstationary hybrid characteristics, Industrial and Engineering Chemistry Research 57: 5045–5057, DOI: 10.1021/acs.iecr.8b00233.
• [47] Zhao, C.H. and Sun, H. (2019). Dynamic distributed monitoring strategy for large-scale nonstationary processes subject to frequently varying conditions under closed-loop control, IEEE Transactions on Industrial Electronics 66(6): 4749–4758, DOI: 10.1109/TIE.2018.2864703.
• [48] Zhirabok, A. and Shumsky, A. (2018). Fault diagnosis in nonlinear hybrid systems, International Journal of Applied Mathematics and Computer Science 28(4): 635–648, DOI: 10.2478/amcs-2018-0049.
• [49] Zou, X.Y. and Zhao, C.H. (2019). Meticulous assessment of operating performance for processes with a hybrid of stationary and nonstationary variables, Industrial and Engineering Chemistry Research 58: 1341–1351, DOI: 10.1021/acs.iecr.8b05005.

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa Nr 461252 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2021).