Improved Mel Frequency Cepstral Coefficients for Compressors and Pumps Fault Diagnosis with Deep Learning Models
Source abstract
Abstract
Compressors and pumps are machines frequently used in petroleum and chemical industries for fluid transportation through flow systems to keep industrial processes running permanently. As their failure can produce costly disruption, developing fault detection and diagnosis tools is essential for accurately detecting and diagnosing faults. This research proposes a bi-dimensional representation of the vibration signal corresponding to the Mel Frequency Cepstral Coefficients (MFCC) and their first two derivatives as features. The pseudo-periodic nature of the fault signature in rotating machines is exploited to put forward an efficient and accurate patch-wise fault classification method. This approach enables the classification of 13 combined types of faults in a multi-stage centrifugal pump and 17 faults in a reciprocating compressor. Classification is performed using the Long Short-Term Memory (LSTM) network, the bidirectional Long Short-Term Memory (BiLSTM) neural network, and the Convolutional Neural Network (CNN). Accurate classification over 99% is attained, showing that the proposed feature extraction procedure correctly classifies a large set of faults simultaneously appearing in such rotating machines.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Machine Fault Diagnosis Techniques · Fault Detection and Control Systems · Power Transformer Diagnostics and Insulation
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
GIDTEC, Universidad Politécnica Salesiana, Cuenca 010105, Ecuador
CIBYTEL—Engineering School, Universidad de Los Andes, Mérida 5101, Venezuela
Mechanical Engineering Department, Universidad de Tarapacá, Arica 1010069, Chile
School of Mechanical Engineering, Dongguan University of Technology, Dongguan 523808, China
Funding
National Natural Science Foundation of China · Universidad Politécnica Salesiana
References (67)
Show source-provided references
1. A fault diagnosis scheme for rotating machinery using hierarchical symbolic analysis and convolutional neural network10.1016/j.isatra.2019.01.018 ↗
2. McKee, K., Forbes, G., Mazhar, M.I., Entwistle, R., and Howard, I. (2024, February 01). A Review of Major Centrifugal Pump Failure Modes with Application To the Water Supply and Sewerage Industries. Available online: http://hdl.handle.net/20.500.11937/28560.
3. Fault detection in reciprocating compressor valves under varying load conditions10.1016/j.ymssp.2015.09.005 ↗
4. Daraz, A., Alabied, S., Smith, A., Gu, F., and Ball, A.D. (2018, January 6–7). Detection and Diagnosis of Centrifugal Pump Bearing Faults Based on the Envelope Analysis of Airborne Sound Signals. Proceedings of the 2018 24th International Conference on Automation and Computing (ICAC), Newcastle Upon Tyne, UK.10.23919/IConAC.2018.8749053 ↗
5. Prediction of flow blockages and impending cavitation in centrifugal pumps using support vector machine (SVM) algorithms based on vibration measurements10.1016/j.measurement.2018.07.092 ↗
6. Fausing Olesen, J., and Shaker, H.R. (2020). Predictive Maintenance for Pump Systems and Thermal Power Plants: State-of-the-Art Review, Trends and Challenges. Sensors, 20.10.3390/s20082425 ↗
7. Bayesian approach and time series dimensionality reduction to LSTM-based model-building for fault diagnosis of a reciprocating compressor10.1016/j.neucom.2019.11.006 ↗
8. A data driven approach for condition monitoring of wind turbine blade using vibration signals through best-first tree algorithm and functional trees algorithm: A comparative study10.1016/j.isatra.2017.02.002 ↗
9. Minimum entropy deconvolution based on simulation-determined band pass filter to detect faults in axial piston pump bearings10.1016/j.isatra.2018.11.040 ↗
10. Experimental investigation and multi-conditions identification method of centrifugal pump using Fisher discriminant ratio and support vector machine10.1177/1687814019878041 ↗
11. Cerrada, M., Macancela, J.C., Cabrera, D., Estupiñan, E., Sánchez, R.V., and Medina, R. (2020). Reciprocating Compressor Multi-Fault Classification Using Symbolic Dynamics and Complex Correlation Measure. Appl. Sci., 10.10.3390/app10072512 ↗
12. Fault Diagnosis of Air Compressor in Nuclear Power Plant Based on Vibration Observation Window10.1109/ACCESS.2020.3043398 ↗
13. Improving accuracy of cavitation severity detection in centrifugal pumps using a hybrid feature selection technique10.1016/j.measurement.2017.05.020 ↗
14. Online ANN-based fault diagnosis implementation using an FPGA: Application in the EFI system of a vehicle10.1016/j.isatra.2019.11.003 ↗
15. Orrù, P.F., Zoccheddu, A., Sassu, L., Mattia, C., Cozza, R., and Arena, S. (2020). Machine Learning Approach Using MLP and SVM Algorithms for the Fault Prediction of a Centrifugal Pump in the Oil and Gas Industry. Sustainability, 12.10.3390/su12114776 ↗
16. Oscillatory behavior-based wavelet decomposition for the monitoring of bearing condition in centrifugal pumps10.1177/1350650117727976 ↗
17. Deep learning and its applications to machine health monitoring10.1016/j.ymssp.2018.05.050 ↗
18. Intelligent fault identification of hydraulic pump using deep adaptive normalized CNN and synchrosqueezed wavelet transform10.1016/j.ress.2022.108560 ↗
19. A data indicator-based deep belief networks to detect multiple faults in axial piston pumps10.1016/j.ymssp.2018.04.038 ↗
20. Jiang, W., Li, Z., Li, J., Zhu, Y., and Zhang, P. (2019). Study on a Fault Identification Method of the Hydraulic Pump Based on a Combination of Voiceprint Characteristics and Extreme Learning Machine. Processes, 7.10.3390/pr7120894 ↗
21. Challenges and opportunities of deep learning models for machinery fault detection and diagnosis: A review10.1109/ACCESS.2019.2938227 ↗
22. Cavitation intensity recognition for high-speed axial piston pumps using 1-D convolutional neural networks with multi-channel inputs of vibration signals10.1016/j.aej.2020.07.052 ↗
23. An adaptive data fusion strategy for fault diagnosis based on the convolutional neural network10.1016/j.measurement.2020.108122 ↗
24. Fault diagnosis of reciprocating compressor based on convolutional neural networks with multisource raw vibration signals10.1155/2019/6921975 ↗
25. Fault diagnosis on slipper abrasion of axial piston pump based on extreme learning machine10.1016/j.measurement.2018.03.050 ↗
26. Improved deep convolution neural network (CNN) for the identification of defects in the centrifugal pump using acoustic images10.1016/j.apacoust.2020.107399 ↗
27. An integrated approach based on improved CEEMDAN and LSTM deep learning neural network for fault diagnosis of reciprocating pump10.1109/ACCESS.2021.3056437 ↗
28. Sequential fault diagnosis based on LSTM neural network10.1109/ACCESS.2018.2794765 ↗
29. Xiao, S., Nie, A., Zhang, Z., Liu, S., Song, M., and Zhang, H. (2020). Fault Diagnosis of a Reciprocating Compressor Air Valve Based on Deep Learning. Appl. Sci., 10.10.3390/app10186596 ↗
30. OrbitNet: A new CNN model for automatic fault diagnostics of turbomachines10.1016/j.asoc.2021.107702 ↗
31. Liu, Y., Duan, L., Yuan, Z., Wang, N., and Zhao, J. (2019). An Intelligent Fault Diagnosis Method for Reciprocating Compressors Based on LMD and SDAE. Sensors, 19.10.3390/s19051041 ↗
32. LSTM: A search space odyssey10.1109/TNNLS.2016.2582924 ↗
33. Medina, R., Cerrada, M., Cabrera, D., Sánchez, R.V., Li, C., and De Oliveira, J.V. (2019, January 3–5). Deep Learning-Based Gear Pitting Severity Assessment Using Acoustic Emission, Vibration and Currents Signals. Proceedings of the 2019 Prognostics and System Health Management Conference (PHM-Paris), Paris, France.10.1109/PHM-Paris.2019.00042 ↗
34. Framewise phoneme classification with bidirectional LSTM and other neural network architectures10.1016/j.neunet.2005.06.042 ↗
35. Gradient-based learning applied to document recognition10.1109/5.726791 ↗
36. Mel Frequency Cepstral Coefficient and its applications: A Review10.1109/ACCESS.2022.3223444 ↗
37. Martin, N., and Mailhes, C. (2024, February 01). About Periodicity and Signal to Noise Ratio-the Strength of the Autocorrelation Function. Available online: https://hal.science/hal-00449085/.
38. Brigato, L., and Iocchi, L. (2021, January 10–15). A close look at deep learning with small data. Proceedings of the 2020 25th International Conference on Pattern Recognition (ICPR), Milan, Italy.10.1109/ICPR48806.2021.9412492 ↗
39. Bearing fault classification based on convolutional neural network in noise environment10.1109/ACCESS.2019.2919126 ↗
40. STFT spectrogram based hybrid evaluation method for rotating machine transient vibration analysis10.1016/j.ymssp.2020.107583 ↗
41. Podder, P., Khan, T.Z., Khan, M.H., and Rahman, M.M. (2014). Comparative performance analysis of hamming, hanning and blackman window. Int. J. Comput. Appl., 96.10.5120/16891-6927 ↗
42. Deep scattering spectrum10.1109/TSP.2014.2326991 ↗
43. Acoustic signal based fault detection on belt conveyor idlers using machine learning10.1016/j.apt.2020.04.034 ↗
44. Lee, J., Choi, H., Park, D., Chung, Y., Kim, H.Y., and Yoon, S. (2016). Fault detection and diagnosis of railway point machines by sound analysis. Sensors, 16.10.3390/s16040549 ↗
45. Sigurdsson, S., Petersen, K.B., and Lehn-Schiøler, T. (2024, February 16). Mel Frequency Cepstral Coefficients: An Evaluation of Robustness of MP3 Encoded Music. Available online: https://archives.ismir.net/ismir2006/paper/000080.pdf.
46. Zhao, X., and Wang, D. (2013, January 26–31). Analyzing noise robustness of MFCC and GFCC features in speaker identification. Proceedings of the 2013 IEEE International Conference on Acoustics, Speech and Signal Processing, Vancouver, BC, Canada.10.1109/ICASSP.2013.6639061 ↗
47. Ganchev, T., Fakotakis, N., and Kokkinakis, G. (2005, January 17–19). Comparative Evaluation of Various MFCC Implementations on the Speaker Verification Task. Proceedings of the 10th International Conference on Speech and Computer (SPECOM 2005), Patras, Greece.
48. Paliwal, K.K. (2024, February 01). Decorrelated and Liftered Filter-Bank Energies for Robust Speech Recognition. Available online: https://maxwell.ict.griffith.edu.au/spl/publications/.
49. Differentiation of discrete multidimensional signals10.1109/TIP.2004.823819 ↗
50. Mushtaq, S., Islam, M.M.M., and Sohaib, M. (2021). Deep Learning Aided Data-Driven Fault Diagnosis of Rotatory Machine: A Comprehensive Review. Energies, 14.10.3390/en14165150 ↗
51. Abramowitz, M., and Stegun, I.A. (1964). Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables.
52. Streaming Convolutional Neural Networks for End-to-End Learning with Multi-Megapixel Images10.1109/TPAMI.2020.3019563 ↗
53. A Full-Image Full-Resolution End-to-End-Trainable CNN Framework for Image Forgery Detection10.1109/ACCESS.2020.3009877 ↗
54. Papadopoulos, A., Korus, P., and Memon, N. (2024, February 01). Hard-Attention for Scalable Image Classification. Available online: https://proceedings.neurips.cc/paper_files/paper/2021/file/7b7916dd2de56297aa29cccb2bbf48d4-Paper.pdf.
55. Bengio, S., Wallach, H., Larochelle, H., Grauman, K., Cesa-Bianchi, N., and Garnett, R. Modern Neural Networks Generalize on Small Data Sets. Proceedings of the Advances in Neural Information Processing Systems.
56. Forest species recognition using macroscopic images10.1007/s00138-014-0592-7 ↗
57. Deep learning in microscopy image analysis: A survey10.1109/TNNLS.2017.2766168 ↗
58. Balles, L., Romero, J., and Hennig, P. (2016). Coupling adaptive batch sizes with learning rates. arXiv.
59. Sutskever, I., Martens, J., Dahl, G., and Hinton, G. (2024, February 01). On the Importance of Initialization and Momentum in Deep Learning. Available online: https://proceedings.mlr.press/v28/sutskever13.pdf.
60. Duchi, J., Hazan, E., and Singer, Y. (2024, February 01). Adaptive Subgradient Methods for Online Learning and Stochastic Optimization. Available online: https://www.jmlr.org/papers/volume12/duchi11a/duchi11a.pdf.
61. Kingma, D.P., and Ba, J. (2014). Adam: A method for stochastic optimization. arXiv.
62. Wilson, A.C., Roelofs, R., Stern, M., Srebro, N., and Recht, B. (2024, February 01). The Marginal Value of Adaptive Gradient Methods in Machine Learning. Available online: https://proceedings.neurips.cc/paper_files/paper/2017/file/81b3833e2504647f9d794f7d7b9bf341-Paper.pdf.
63. Use of ranks in one-criterion variance analysis10.1080/01621459.1952.10483441 ↗
64. Maaten, L.v.d., and Hinton, G. (2024, February 01). Visualizing Data Using t-SNE. Available online: https://www.jmlr.org/papers/volume9/vandermaaten08a/vandermaaten08a.pdf?fbcl.
65. Jones, G.M., Sanks, R.L., Tchobanoglous, G., and Bosserman, B.E. (2008). Pumping Station Design, Butterworth-Heinemann. [3rd ed.].
66. Analysis of vibration and noise induced by unsteady flow inside a centrifugal compressor10.1016/j.ast.2020.106286 ↗
67. Deep convolutional neural network based on adaptive gradient optimizer for fault detection in SCIM10.1016/j.isatra.2020.10.052 ↗