A deep learning approach for anomaly detection based on SAE and LSTM in mechanical equipment

  • Zhe LiEmail author
  • Jingyue Li
  • Yi Wang
  • Kesheng WangEmail author


Anomaly in mechanical systems may cause equipment to break down with serious safety, environment, and economic impact. Since many mechanical equipment usually operates under tough working environments, which makes them vulnerable to types of faults, anomaly detection for mechanical equipment usually requires considerable domain knowledge. However, a common dilemma in many practical applications is that one may not be able to obtain the empirical knowledge about anomaly or the history data is completely unlabelled, which makes conventional fault identification methods not applicable. In order to fill the gap, this paper proposes a novel deep learning–based method for anomaly detection in mechanical equipment by combining two types of deep learning architectures, stacked autoencoders (SAE) and long short-term memory (LSTM) neural networks, to identify anomaly condition in a completely unsupervised manner. The proposed method focuses on the anomaly detection through multiple features sequence when the history data is unlabelled and the empirical knowledge about anomaly is absent. An experiment for anomaly detection in rotary machinery through wavelet packet decomposition (WPD) and data-driven models demonstrates the efficiency and stability of the proposed approach. The method can be divided into two stages: SAE-based multiple features sequence representation and LSTM-based anomaly identification. During the experiment, fivefold cross-validation has been applied to validate the performance and stability of the proposed approach. The results show that the proposed approach could detect anomaly working condition with 99% accuracy under a completely unsupervised learning environment and offer an alternative method to leverage and integrate features for anomaly detection without empirical knowledge.


Anomaly detection Mechanical equipment SAE LSTM 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


Funding information

The work described in this article has been conducted as part of the research project CIRCit (Circular Economy Integration in the Nordic Industry for Enhanced Sustainability and Competitiveness), which is part of the Nordic Green Growth Research and Innovation Programme (grant number: 83144), and funded by NordForsk, Nordic Energy Research, and Nordic Innovation.


  1. 1.
    Lei Y, Lin J, He Z, Zuo MJ (2013) A review on empirical mode decomposition in fault diagnosis of rotating machinery. Mech Syst Signal Process 35(1):108–126CrossRefGoogle Scholar
  2. 2.
    Hernandez-Vargas M, Cabal-Yepez E, Garcia-Perez A (2014) Real-time SVD-based detection of multiple combined faults in induction motors. Comput Electr Eng 40(7):2193–2203CrossRefGoogle Scholar
  3. 3.
    El Kadiri S, Grabot B, Thoben K-D, Hribernik K, Emmanouilidis C, von Cieminski G et al (2016) Current trends on ICT technologies for enterprise information systems. Comput Ind 79(Supplement C):14–33. CrossRefGoogle Scholar
  4. 4.
    Precup R-E, Angelov P, Costa BSJ, Sayed-Mouchaweh M (2015) An overview on fault diagnosis and nature-inspired optimal control of industrial process applications. Comput Ind 74(Supplement C):75–94. CrossRefGoogle Scholar
  5. 5.
    Wang P, Guo C (2013) Based on the coal mine’s essential safety management system of safety accident cause analysis. Am J Environ Energy Power Res 1(3):62–68Google Scholar
  6. 6.
    Ayele YZ, Barabadi A (2016) Risk based inspection of offshore topsides static mechanical equipment in Arctic conditions. In: 2016 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM). 4–7 Dec. 2016. pp 501–506Google Scholar
  7. 7.
    Gao Z, Cecati C, Ding SX (2015) A survey of fault diagnosis and fault-tolerant techniques—part i: fault diagnosis with model-based and signal-based approaches. IEEE Trans Ind Electron 62(6):3757–3767. CrossRefGoogle Scholar
  8. 8.
    Klingert F, Roeder G, Schellenberger M, Bauer A, Frey L, Brueggemann M et al (2017) Condition-based maintenance of mechanical setup in aluminum wire bonding equipment by data mining. In: Information and Communication Technology, Electronics and Microelectronics (MIPRO), 2017 40th International Convention on. IEEE, pp 72–77Google Scholar
  9. 9.
    Yang Y, Dong X, Peng Z, Zhang W, Meng G (2015) Vibration signal analysis using parameterized time–frequency method for features extraction of varying-speed rotary machinery. J Sound Vib 335:350–366CrossRefGoogle Scholar
  10. 10.
    Bangalore P, Tjernberg LB (2015) An artificial neural network approach for early fault detection of gearbox bearings. IEEE Trans Smart Grid 6(2):980–987. CrossRefGoogle Scholar
  11. 11.
    López-Pérez D, Antonino-Daviu J (2017) Application of infrared thermography to failure detection in industrial induction motors: case stories. IEEE Trans Ind Appl 53(3):1901–1908. CrossRefGoogle Scholar
  12. 12.
    Lin J, Chen Q (2014) A novel method for feature extraction using crossover characteristics of nonlinear data and its application to fault diagnosis of rotary machinery. Mech Syst Signal Process 48(1):174–187CrossRefGoogle Scholar
  13. 13.
    Griffin JM, Doberti AJ, Hernández V, Miranda NA, Vélez MA (2017) Multiple classification of the force and acceleration signals extracted during multiple machine processes: part 1 intelligent classification from an anomaly perspective. Int J Adv Manuf Technol 93(1):811–823. CrossRefGoogle Scholar
  14. 14.
    Lu C, Wang Z-Y, Qin W-L, Ma J (2017) Fault diagnosis of rotary machinery components using a stacked denoising autoencoder-based health state identification. Signal Process 130:377–388CrossRefGoogle Scholar
  15. 15.
    Aydin I, Karakose M, Akin E (2015) Anomaly detection using a modified kernel-based tracking in the pantograph–catenary system. Expert Syst Appl 42(2):938–948. CrossRefGoogle Scholar
  16. 16.
    Li Z, Wang Y, Wang K (2017) A data-driven method based on deep belief networks for backlash error prediction in machining centers. J Intell Manuf.
  17. 17.
    Peña M, Biscarri F, Guerrero JI, Monedero I, León C (2016) Rule-based system to detect energy efficiency anomalies in smart buildings, a data mining approach. Expert Syst Appl 56:242–255. CrossRefGoogle Scholar
  18. 18.
    Zhou Q, Yan P, Liu H, Xin Y, Chen Y (2018) Research on a configurable method for fault diagnosis knowledge of machine tools and its application. Int J Adv Manuf Technol 95(1):937–960. CrossRefGoogle Scholar
  19. 19.
    Diez-Olivan A, Pagan JA, Khoa NLD, Sanz R, Sierra B (2018) Kernel-based support vector machines for automated health status assessment in monitoring sensor data. Int J Adv Manuf Technol 95(1):327–340. CrossRefGoogle Scholar
  20. 20.
    Landry M, Leonard F, Landry C, Beauchemin R, Turcotte O, Brikci F (2008) An improved vibration analysis algorithm as a diagnostic tool for detecting mechanical anomalies on power circuit breakers. IEEE Trans Power Deliv 23(4):1986–1994. CrossRefGoogle Scholar
  21. 21.
    Amruthnath N, Gupta T (2018) A research study on unsupervised machine learning algorithms for early fault detection in predictive maintenance. In: 2018 5th International Conference on Industrial Engineering and Applications (ICIEA). IEEE, pp 355–361Google Scholar
  22. 22.
    von Birgelen A, Buratti D, Mager J, Niggemann O (2018) Self-organizing maps for anomaly localization and predictive maintenance in cyber-physical production systems. Proced CIRP 72:480–485. CrossRefGoogle Scholar
  23. 23.
    Costa BSJ, Angelov PP, Guedes LA (2015) Fully unsupervised fault detection and identification based on recursive density estimation and self-evolving cloud-based classifier. Neurocomputing 150:289–303. CrossRefGoogle Scholar
  24. 24.
    Serdio F, Lughofer E, Pichler K, Buchegger T, Efendic H (2014) Residual-based fault detection using soft computing techniques for condition monitoring at rolling mills. Inf Sci 259:304–320. CrossRefGoogle Scholar
  25. 25.
    Hu Q, He Z, Zhang Z, Zi Y (2007) Fault diagnosis of rotating machinery based on improved wavelet package transform and SVMs ensemble. Mech Syst Signal Process 21(2):688–705. CrossRefGoogle Scholar
  26. 26.
    Zhang Y, Liu B, Ji X, Huang DJNPL (2017) Classification of EEG signals based on autoregressive model and wavelet packet decomposition (journal article). 45(2):365–378.
  27. 27.
    Ferreira CBR, Borges D b L (2003) Analysis of mammogram classification using a wavelet transform decomposition. Pattern Recogn Lett 24(7):973–982CrossRefGoogle Scholar
  28. 28.
    Murugappan M, Ramachandran N, Sazali Y (2010) Classification of human emotion from EEG using discrete wavelet transform. J Biomed Sci Eng 3(04):390–396CrossRefGoogle Scholar
  29. 29.
    Rastbood A, Majdi A, Gholipour Y (2017) Prediction of structural forces of segmental tunnel lining using FEM based artificial neural network. Int J Min Geo-Eng 51(1):71–78Google Scholar
  30. 30.
    Bengio Y, Lamblin P, Popovici D, Larochelle H (2007) Greedy layer-wise training of deep networks. Adv Neural inform Process Syst 153–160Google Scholar
  31. 31.
    Poultney C, Chopra S, Cun YL (2007) Efficient learning of sparse representations with an energy-based model. In: Advances in neural information processing systems. pp 1137–1144Google Scholar
  32. 32.
    Bengio Y, Courville A, Vincent P (2013) Representation learning: a review and new perspectives. IEEE Trans Pattern Anal Mach Intell 35(8):1798–1828CrossRefGoogle Scholar
  33. 33.
    Deng L (2012). Three classes of deep learning architectures and their applications: a tutorial survey. In: APSIPA transactions on signal and information processingGoogle Scholar
  34. 34.
    Galloway GS, Catterson VM, Fay T, Robb A, Love C (2016) Diagnosis of tidal turbine vibration data through deep neural networks. In: Proceedings of the Third European Conference of the Prognostics and Health Management Society 2016. (PHM Society), pp 172–180Google Scholar
  35. 35.
    Erhan D, Bengio Y, Courville A, Manzagol P-A, Vincent P, Bengio S (2010) Why does unsupervised pre-training help deep learning? J Mach Learn Res 11(Feb):625–660MathSciNetzbMATHGoogle Scholar
  36. 36.
    Jiang S, Chin K-S, Wang L, Qu G, Tsui KL (2017) Modified genetic algorithm-based feature selection combined with pre-trained deep neural network for demand forecasting in outpatient department. Expert Syst Appl 82:216–230. CrossRefGoogle Scholar
  37. 37.
    Cortez B, Carrera B, Kim Y-J, Jung J-Y (2018) An architecture for emergency event prediction using LSTM recurrent neural networks. Expert Syst Appl 97:315–324. CrossRefGoogle Scholar
  38. 38.
    Hochreiter S, Schmidhuber J (1997) Long short-term memory. Neural Comput 9(8):1735–1780. CrossRefGoogle Scholar
  39. 39.
    Schmidhuber J (2015) Deep learning in neural networks: an overview. Neural Netw 61:85–117CrossRefGoogle Scholar
  40. 40.
    Greff K, Srivastava RK, Koutník J, Steunebrink BR, Schmidhuber J (2016) LSTM: a search space odyssey. IEEE Trans Neural Netw Learn SystGoogle Scholar
  41. 41.
    Ma X, Tao Z, Wang Y, Yu H, Wang Y (2015) Long short-term memory neural network for traffic speed prediction using remote microwave sensor data. Transp Res C Emerg Technol 54:187–197CrossRefGoogle Scholar
  42. 42.
    Zhao R, Yan R, Wang J, Mao K (2017) Learning to monitor machine health with convolutional bi-directional lstm networks. Sensors 17(2):273CrossRefGoogle Scholar
  43. 43.
    Liao L, Ahn H-I (2016) Combining deep learning and survival analysis for asset health management. Int J Prognostics Health ManagGoogle Scholar
  44. 44.
    Sak H, Senior A, Beaufays F (2014) Long short-term memory recurrent neural network architectures for large scale acoustic modeling. Fifteenth Ann Conf Int Speech Commun AssocGoogle Scholar
  45. 45.
    Sutskever I, Vinyals O, Le QV (2014) Sequence to sequence learning with neural networks. Advances in neural information processing systems, pp 3104–3112Google Scholar
  46. 46.
    de Bruin T, Verbert K, Babuška R (2017) Railway track circuit fault diagnosis using recurrent neural networks. IEEE Trans Neural Netw Learn Syst 28(3):523–533MathSciNetCrossRefGoogle Scholar
  47. 47.
    Zhuge Q, Xu L, Zhang G (2017) LSTM neural network with emotional analysis for prediction of stock price. Eng Lett 25(2):167–175Google Scholar
  48. 48.
    Cheng S, Pecht M (2012) Using cross-validation for model parameter selection of sequential probability ratio test. Expert Syst Appl 39(9):8467–8473. CrossRefGoogle Scholar
  49. 49.
    Li Z, Wang Y, Wang K (2019) A deep learning driven method for fault classification and degradation assessment in mechanical equipment. Comput Ind 104:1–10. CrossRefGoogle Scholar
  50. 50.
    Zamir AR, Sax A, Shen W, Guibas LJ, Malik J, Savarese S (2018) Taskonomy: disentangling task transfer learning. Proceed IEEE Conf Comput Vision Pattern Recogn 3712–3722Google Scholar

Copyright information

© Springer-Verlag London Ltd., part of Springer Nature 2019

Authors and Affiliations

  1. 1.Department of Computer ScienceNorwegian University of Science and TechnologyTrondheimNorway
  2. 2.School of BusinessPlymouth UniversityPlymouthUK
  3. 3.School of Mechanical EngineeringChangzhou UniversityChangzhouChina
  4. 4.Department of Mechanical and Industrial EngineeringNorwegian University of Science and TechnologyTrondheimNorway

Personalised recommendations