Diagnóstico de fallas mediante una LSTM y una red elástica

M. A. Márquez-Vera, O. López-Ortega, L. E. Ramos-Velasco, R. M. Ortega-Mendoza, B. J. Fernández-Neri, N. S. Zúñiga-Peña


El diagnóstico de fallas es importante en los procesos industriales, ya que permite determinar si es necesario detener el proceso en operación y/o proponer un plan de mantenimiento. En el presente trabajo se comparan dos estrategias para diagnosticar fallas. La primera realiza un preprocesamiento de datos usando el análisis de componentes independientes para reducir la dimensión de los datos, posteriormente, se emplea la transformada wavelet para resaltar las señales de falla, con esta información se alimenta una red neuronal artificial. Por su parte, la segunda estrategia, principal contribución de este trabajo, usa una memoria de corto y largo plazo. Esta memoria es alimentada por las variables más significativas seleccionadas mediante una red elástica para usar tanto la norma $L_1$ como la $L_2$. Como ejemplo de aplicación se utilizó el proceso químico Tennessee Eastman, un proceso ampliamente usado en el diagnóstico de fallas. El aislamiento de fallas mostró mejores resultados con respecto a los reportados en la literatura.

Palabras clave

Diagnóstico de fallas; Transformada Wavelet; Redes neuronales recurrentes; Análisis de componentes independientes; Red elástica

Clasificación por materias

Detección, aislamiento, diagnóstico, identificación, estimación y acomodación de fallos; transformada wavelet; control neuronal

Texto completo:



Adewole, A., Tzoneva, R., Behardien, S., 2016. Distribution network fault section identification and fault location using wavelet entropy and neural networks. Applied Soft Computing 46, 296-306. https://doi.org/10.1016/j.asoc.2016.05.013

Alkaya, A., Eker, I., 2011. Variance sensitive adaptive threshold-based PCA method for fault detection with experimental application. ISA Transactions 50, 287-302. https://doi.org/10.1016/j.isatra.2010.12.004

Barakat, S., Eteiba, M., Wahba, W., 2014. Fault location in underground cables using anfis nets and discrete wavelet transform. Journal of Electrical Systems and Information Technology 1, 198-211. https://doi.org/10.1016/j.jesit.2014.12.003

Bathelt, A., Ricker, N., Jelali, M., 2015. Revision of the Tennessee Eastman process model. IFAC Papers-Online 48 (8), 309-314. https://doi.org/10.1016/j.ifacol.2015.08.199

Boldt, F., Rauber, T., Varejao, F., October 2014. Evaluation of the extreme learning machine for automatic fault diagnosis of the Tennessee Eastman chemical process. In: IEEE (Ed.), Annual Conference of the IEEE Industrial Electronics Society. Vol. 40. Dallas, Texas, pp. 2551-2557. https://doi.org/10.1109/IECON.2014.7048865

Chen, H., Tino, P., Yao, X., 2014. Cognitive fault diagnosis in Tennessee Eastman process using learning in the model space. Computers and Chemical Engineering 67, 33-42. https://doi.org/10.1016/j.compchemeng.2014.03.015

Rodrigues, J., Filho, P., PeixotoJr., E., Kumar, A., deAlbuquerque, V., 2019. Classification of EEG signals to detect alcoholism using machine learning techniques. Pattern Recognition Letters 125, 140-149. https://doi.org/10.1016/j.patrec.2019.04.019

Dixit, A., Majumdar, S., 2013. Comparative analysis of coiflet and daubechies wavelets using global threshold for image denoising. Intenational Journal of Advances in Engineering & Technology 6 (5), 2247-2252.

Downs, J., Vogel, E., 1993. A plant-wide industrial process control problem. Computers and Chemical Engineering 17 (3), 245-255. https://doi.org/10.1016/0098-1354(93)80018-I

Fischer, T., Krauss, C., 2018. Deep learning with long short-term memory networks for financial market predictions. European Journal of Operational Research 270, 654-669. https://doi.org/10.1016/j.ejor.2017.11.054

Gao, X., Hou, J., 2016. An improved SVM integrated GS-PCA fault diagnosis approach of Tennessee Eastman process. Neurocomputing 174, 906-911. https://doi.org/10.1016/j.neucom.2015.10.018

Geng, Z., Li, Z., Han, Y., 2018. A new deep belief network based on RBM with glial chains. Information Sciences 463, 294-306. https://doi.org/10.1016/j.ins.2018.06.043

Goodfellow, I., Bengio, Y., Courville, A., 2016. Deep Learning. MIT Press, United States of America, http://www.deeplearningbook.ogr.

Han, L., Li, C., Guo, S., Su, X., 2015. Feature extraction method of bearing AE signal based on improved Fast-ICA and wavelet packet energy. Mechanical Systems and Signal Processing 62-63, 91-99. https://doi.org/10.1016/j.ymssp.2015.03.009

Hastie, T., Tibshirani, R., Friedman, J., 2009. The elements of statistical learning: data mining, inference and prediction. Springer, New York. https://doi.org/10.1007/978-0-387-84858-7

Hoang, D., Kang, H., 2019. A survey on deep learning based bearing fault diagnosis. Neurocomputing 335, 327-335. ttps://doi.org/10.1016/j.neucom.2018.06.078

Hochreiter, S., Schmidhuber, J., 1997. Long short term memory. Neural Computation 9 (8), 1735-1780. ttps://doi.org/10.1162/neco.1997.9.8.1735

Hyvärinen, A., Oja, E., 2000. Independent component analysis: Algorithms and applications. Neural Networks 13, 411-430. ttps://doi.org/10.1016/S0893-6080(00)00026-5

Jing, C., Gao, X., Zhu, X., Lang, S., July 2014. Fault classificaction on Tennessee Eastman process: PCA and SVM. In: IEEE (Ed.), Intenational Conference on Mecatronics and Control. Jinzhou, China, pp. 2194-2197. https://doi.org/10.1109/ICMC.2014.7231958

Jung, C., Kim, K., Lee, J., Klockl, B., 2007. Wavelet and neuro-fuzzy based fault location for combined transmission systems. Energy Systems 29, 445-454. https://doi.org/10.1016/j.ijepes.2006.11.003

Kandula, V. K., 2011. Fault detection in process control plants using principal component analysis. Master's thesis, Louisiana State University, Department of Electrical Engineering.

Karpenko, M., Sepehri, N., Octubre 2001. A neural network based fault detection and identification scheme for pneumatic process control valves. In: IEEE (Ed.), International Conference on Systems, Man and Cybernetics. Tucson, USA, pp. 93-98. https://doi.org/10.1109/ICSMC.2001.969794

Khakipour, M., Safavi, A., Setoodeh, P., 2017. Bearing fault diagnosis with morphological gradient wavelet. Journal of the Franklin Institute 354, 2465-2476. https://doi.org/10.1016/j.jfranklin.2016.11.013

Kuang, T., Yang, Z., Yao, Y., 2015. Multivariate fault isolation via variable selection in discriminant analysis. Journal of Process Control 35, 30-40. https://doi.org/10.1016/j.isatra.2017.06.014

Kumar, R., Bansal, H., 2019. Hardware in the loop implementation of wavelet based strategy in shuntactive powerfilter to mitigate power quality issues. Electric Power Systems Research 169, 92-104. https://doi.org/10.1016/j.epsr.2019.01.001

Lau, C., Ghosh, K., Hussain, M., Hassan, C. C., 2013. Fault disgnosis of Tennessee Eastman process with multi-scale PCA and ANFIS. Chemom. Intell. Lab. Syst. 120, 1-14. https://doi.org/10.1016/j.chemolab.2012.10.005

Lee, J., Yoo, C., Lee, I., 2004. Statistical process monitoring with independent component analysis. Journal of Process Control 14 (5), 467-485. https://doi.org/10.1016/j.jprocont.2003.09.004

Lei, J., Liu, C., Jiang, D., 2019. Fault diagnosis of wind turbine based on long short-term memory networks. Renewable Energy 133, 422-432. https://doi.org/10.1016/j.renene.2018.10.031

Li, W., Monti, A., Ponci, F., 2014. Fault detection and classification in medium voltage DC shipboard power systems with wavelets and artificial neural networks. IEEE Transactions on Instrumentation and Measurement 63 (11), 2651-2665. https://doi.org/10.1109/TIM.2014.2313035

Liang, P., Deng, C.,Wu, J., Yang, Z., Zhu, J., Zhang, Z., 2019. Compound fault diagnosis of gearboxes via multi-label convolutional neural network and wavelet transform. Computers in Industry 113, 103132. https://doi.org/10.1016/j.compind.2019.103132

Lin, J., Zhang, A., 2005. Fault feature separation using wavelet-ICA filter. NDT&E International 38, 421-427. https://doi.org/10.1016/j.ndteint.2004.11.005

Linker, R., Gutman, P., Seginer, I., 2002. Observer-based robust failure detection and isolation in greenhouses. Control Engineering Practice 10 (5), 519- 531. https://doi.org/10.1016/S0967-0661(02)00002-3

Lou, W., Loparo, K., 2004. Bearing fault diagnosis based on wavelet transform and fuzzy inference. Mechanical Systems and Signal Processing 18, 1077-1095. https://doi.org/10.1016/S0888-3270(03)00077-3

Lv, F.,Wen, C., Bao, Z., Liu, M., 2016. Fault diagnosis based on deep learning. In: AACC (Ed.), American Control Conference. Boston, USA, pp. 6851-6856. https://doi.org/10.1109/ACC.2016.7526751

Lv, F., Wen, C., Liu, M., Bao, Z., 2017. Weighted time series fault diagnosis based on a staked sparce autoencoder. Journal of Chemometrics 31, 16 pages. https://doi.org/10.1002/cem.2912

Lv, F., Fan, X., Wen, C., Bao, Z., 2018. Stacked sparse auto encoder network based multimode process monitoring. In: IEEE (Ed.), International Conference on Control Automation & Information Science. Hangzhou, China, pp. 227-232. https://doi.org/10.1109/ICCAIS.2018.8570618

Maglaveras, N., Stamkopoulos, T., Diamantaras, K., Pappas, C., Strintzis, M., 1998. ECG pattern recognition and classification using non-linear transfor mations and neural networks: A review. International Journal of Medical Informatics 52, 191-208. https://doi.org/10.1016/S1386-5056(98)00138-5

Methnani, S., Lafont, F., Gautier, J., Damak, T., Toumi, A., 2013. Actuator and sensor fault detection, isolation and identification in nonlinear dynamical systems, with applications to a waste water treatment plant. Journal of Computer Engineering and Informatics 1 (4), 112-125. https://doi.org/10.1080/21642583.2014.888525

Muñoz-Cobo, J., Mendizábal, R., Miquel, A., Berna, C., Escrivá, A., 2017. Use of the principles of maximum entropy and maximum relative entropy for the determination of uncertain parameter distributions in engineering applications. Entropy 19, 486, 37 pages. https://doi.org/10.3390/e19090486

Nguyen, B., Quyen, A., Nguyen, P., Ton, T., July 2017. Wavelet-based neural network for recognition of faults at nhabe power substation of the vietnam power system. In: IEEE (Ed.), International Conference on System Science and Engineering. Ho Chi Minh City, Vietnam, pp. 165-168. https://doi.org/10.1109/ICSSE.2017.8030858

Ojeda-González, A., Mendes-Jr., O., Oliveira-Domingues, M., Menconi, V., 2014. Daubechies wavelet coeffcients: a tool to study interplanetary magnetic field fluctuations. Geof'ısica Internacional 53 (2), 101-115. https://doi.org/10.1016/S0016-7169(14)71494-1

Oliveira, J., Pontes, K., Santori, I., Embirucu, M., 2017. Fault detection and diagnosis in dynamic systems using weightless neural networks. Expert Systems With Applications 84, 200-219. https://doi.org/10.1016/j.eswa.2017.05.020

Patan, K., 2008. Artificial neural networks for the modelling and fault diagnosis of technical process. Lecture Notes in Control and Information Sciences. Springer, India.

Rafiee, J., Rafiee, M., Tse, P., 2010. Application of mother wavelet functions for automatic gear and bearing fault diagnosis. Expert Systems with Applications 37, 4568-4579. https://doi.org/10.1016/j.eswa.2009.12.051

Ramos-Velasco, L., Ramos-Fernández, J., Islar-Gómez, O., Espejel-Rivera, M., García-Lamont, J., Márquez-Vera, M., 2013. Identificación y control wavenet de un motor de ca. Revista Iberoamericana de Automática e Informática Industrial 10, 269-278. https://doi.org/10.1016/j.riai.2013.05.002

Rato, T., Reis, M., 2013. Defining the structure of DPCA models and its impact on process monitoring and prediction ctivities. Chemometrics and Intelligent Laboratory Systems 125, 74-86. https://doi.org/10.1016/j.chemolab.2013.03.009

Rockinger, M., Jondeau, E., 2002. Entropy densities with an application to autoregressive conditional skewness and kurtosis. Journal of Econometrics 106, 119-142. https://doi.org/10.1016/S0304-4076(01)00092-6

Salahschoor, K., Kiasi, F., July 2008. On-line process monitoring based on wavelet-ICA methodology. In: IFAC (Ed.), Proceedings of the 17th World Congress. Seul- Korea, pp. 6-11. https://doi.org/10.3182/20080706-5-KR-1001.01253

Salahshoor, K., Khoshro, M., Kordestani, M., 2011. Fault detection and diagnosis of an industrial steam turbine using a distributed configuration of adaptive neuro-fuzzy inference systems. Simulation Modelling Practice and Theory 19, 1280-1293. https://doi.org/10.1016/j.simpat.2011.01.005

Sharif, I., Khare, S., 2014. Comparative analysis of Haar and Daubechies wavelet for hyper spectral image classification. In: Commission, I. T. (Ed.), VIII Symposium of The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Science. Hyderabad-India, pp. 937-941. https://doi.org/10.5194/isprsarchives-XL-8-937-2014

Smirnov, E., Timoshenko, D., Adrianov, S., 2014. Comparison of regularization methods for imagenet classification with deep convolutional neural networks. AASRI Procedia 6, 89-94. https://doi.org/10.1016/j.aasri.2014.05.013

Sobhani-Tehrani, E., Khorasani, K., 2009. Fault diagnosis of nonlinear systems using a hybrid approach. Fault detetion and diagnosis. Springer, Berlin, Ch. 2, pp. 22-49. https://doi.org/10.1007/978-0-387-92907-1_2

Tayarani-Bathaie, S., Vanini, Z., Khorasani, K., 2014. Dynamic neural networkbased fault diagnosis of gas turbine engines. Neurocomputing 125, 153-165. https://doi.org/10.1016/j.neucom.2012.06.050

Zvokelj, M., Zupan, S., Prebil, I., 2016. EEMD-based multiscale ICA method for slewing bearing fault detection and diagnosis. Journal of Sound and Vibration 26, 394-423. https://doi.org/10.1016/j.jsv.2016.01.046

Wang, X., Qin, Y., Wang, Y., Xiang, S., Chen, H., 2019. ReLTanh: An activation function with vanishing gradient resistance for SAE-based DNNs and its application to rotating machinery fault diagnosis. Neurocomputing 363, 88-98. https://doi.org/10.1016/j.neucom.2019.07.017

Wu, F., Tong, F., Yang, Z., 2016. EMGdi signal enhancement based on ICA decomposition and wavelet transform. Applied Soft Computing 43, 561-571. https://doi.org/10.1016/j.asoc.2016.03.002

Wu, J., Hsu, C., Wu, G., 2009. Fault gear identification and classification using discrete wavelet transform and adaptive neuro-fuzzy inference. Expert Systems with Applications 36, 6244-6255. https://doi.org/10.1016/j.eswa.2008.07.023

Wu, Q., Law, R., Wu, S., 2011. Fault diagnosis of car assembly line based on fuzzy wavelet kernel support vector classifier machine and modified genetic algorithm. Expert Systems with Applications 38, 9096-9104. https://doi.org/10.1016/j.eswa.2010.12.109

Wu, H., Zhao., Jinsong, 2018. Deep convolutional neural network model based chemical process fault diagnosis. Computers and Chemical Engineering 115, 185-197. https://doi.org/10.1016/j.compchemeng.2018.04.009

Xiao, C., Chen, N., Hu, C., Wang, K., Gong, J., Chen, Z., 2019. Short and midterm sea surface temperature prediction using time-series satellite data and LSTM-AdaBoost combination approach. Remote Sensing of Environment 233, 111358. https://doi.org/10.1016/j.rse.2019.111358

Xie, D., Bai, L., December 2015. A hierarchical deep neural network for fault diagnosis on Tennessee-Eastman process. In: IEEE (Ed.), International Conference on Machine Learning and Applications. Vol. 14. Miami, USA, pp. 745-748. https://doi.org/10.1109/ICMLA.2015.208

Yan, R., Gao, R., Chen, X., 2014. Wavelets for fault diagnosis of rotary machines: A review with applications. Signal Processing 351, 4555-4569. https://doi.org/10.1016/j.sigpro.2013.04.015

Yan, Z., Yao, Y., 2015. Variable selection method for fault isolation using least absolute shrinkage and selection operator (LASSO). Chemometrics and Intelligent Laboratory Systems 146, 136-146. https://doi.org/10.1016/j.chemolab.2015.05.019

Yao, G., Lei, T., Zhong, J., 2019. A review of convolutional-neural-networkbased action recognition. Pattern Recognition Letters 118, 14-22. https://doi.org/10.1016/j.patrec.2018.05.018

Yin, S., Ding, S., Haghani, A., Hao, H., 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, 1567-1581. https://doi.org/10.1016/j.jprocont.2012.06.009

Zhang, Q., Yang, L., Chen, Z., Li, P., 2018. A survey on deep learning for big data. Information Fusion 42, 146-157. https://doi.org/10.1016/j.inffus.2017.10.006

Zhang, X., Polycarpou, M., Parisini, T., 2002. A robust detection and isolation scheme for abrupt and incipient faults in nonlinear systems. IEEE Transactions on Automatic Control 47 (4), 576-593. https://doi.org/10.1109/9.995036

Zhang, Y., Zhang, L., Zhang, H., 2012. Fault detection for industrial processes. Mathematical Problems in Engineering 2012, 18 pages. https://doi.org/10.1155/2012/757828

Zhang, Z., Zhao, J., 2017. A deep belief network based fault diagnosis model for complex chemical process. Computers and Chemical Engineering 107, 395-407. https://doi.org/10.1016/j.compchemeng.2017.02.041

Zhao, H., 2018. Neural component analysis for fault detection. Chemometrics and Intelligent Laboratory Systems 176, 11-21. https://doi.org/10.1016/j.chemolab.2018.02.001

Zhao, R., Yan, R., Chen, Z., Mao, K., Wang, P., Gao, R., 2019. Deep learning and its applications to machine health monitoring. Mechanical Systems and Signal Processing 115, 213-237. https://doi.org/10.1016/j.ymssp.2018.05.050

Zheng, J., Huang, W., Wang, Z., Liang, J., 2019. Mutual information-based sparse multiblock dissimilarity method for incipient fault detection and diagnosis in plant-wide process. Journal of Process Control 83, 63-76. https://doi.org/10.1016/j.jprocont.2019.09.004

Abstract Views

Metrics Loading ...

Metrics powered by PLOS ALM

Creative Commons License

Esta revista se publica bajo una Licencia Creative Commons Attribution-NonCommercial-CompartirIgual 4.0 International (CC BY-NC-SA 4.0)

Universitat Politècnica de València     https://doi.org/10.4995/riai

e-ISSN: 1697-7920     ISSN: 1697-7912