A fuzzy-neuro approach to predictive maintenance using vibration signature classification
-
Asokan Vasudevan
Faculty of Business and Communications, INTI International University, Nilai 71800, Malaysia
-
Mohammed El Khider
Department of General Undergraduate Curriculum Requirements, University of Dubai, Dubai P.O. Box 14143, United Arab Emirates
-
Yogeesh N
Faculty of Business and Communications, INTI International University, Nilai 71800, Malaysia; Department of Mathematics, Government First Grade College, Tumakuru 572102, India
-
Puspanathan Doraisingam
Faculty of Business and Communications, INTI International University, Nilai 71800, Malaysia
-
Khan Sarfaraz Ali
Faculty of Business and Communications, INTI International University, Nilai 71800, Malaysia
-
A. Sathishkumar
Department of Biomedical Engineering, Erode Sengunthar Engineering College, Perundurai 638057, India
Abstract
Vibration-based predictive maintenance must remain reliable under variable speed/load and noise while delivering actionable, interpretable decisions. We propose a fuzzy-neuro framework that maps windowed vibration segments to class decisions and a calibrated risk score with uncertainty. Given a sampled signal , overlapping windows are formed and encoded into a learned signature . Physics-informed indicators (e.g., RMS, kurtosis, band-energy and short-horizon trend) are fused with to form . An end-to-end neuro-adaptive Takagi-Sugeno-Kang (TSK) layer produces a transparent maintenance risk enabling rule-level explanations via dominant firing strengths . To support safe decision-making near thresholds, we estimate an interval risk from predictive samples using quantiles , and width as confidence. Using leakage-safe, unit-aware temporal splitting, experiments on public rolling-bearing benchmarks achieve 0.956 accuracy, 0.952 macro-F1, and 0.941 MCC, while the full model improves calibration and yields sharper risk intervals (mean width 0.132), translating classifier evidence into auditable "monitor/schedule/urgent" actions. These results indicate that the proposed framework is accurate, interpretable, and decision-ready for predictive maintenance.
Copyright (c) 2026 Asokan Vasudevan, Mohammed El Khider, Yogeesh N, Puspanathan Doraisingam, Khan Sarfaraz Ali, A. Sathishkumar
This work is licensed under a Creative Commons Attribution 4.0 International License.
References
[1]Zadeh LA. Fuzzy sets. Information and Control. 1965; 8(3): 338–353. doi: 10.1016/S0019-9958(65)90241-X
[2]Takagi T, Sugeno M. Fuzzy identification of systems and its applications to modeling and control. IEEE Transactions on Systems, Man, and Cybernetics. 1985; SMC-15(1): 116–132. doi: 10.1109/TSMC.1985.6313399
[3]Jang JSR. ANFIS: Adaptive-network-based fuzzy inference system. IEEE Transactions on Systems, Man, and Cybernetics. 1993; 23(3): 665–685. doi: 10.1109/21.256541
[4]Randall RB, Antoni J. Rolling element bearing diagnostics—A tutorial. Mechanical Systems and Signal Processing. 2011; 25(2): 485–520. doi: 10.1016/j.ymssp.2010.07.017
[5]Antoni J. The spectral kurtosis: A useful tool for characterising non-stationary signals. Mechanical Systems and Signal Processing. 2006; 20(2): 282–307. doi: 10.1016/j.ymssp.2004.09.001
[6]Antoni J. Fast computation of the kurtogram for the detection of transient faults. Mechanical Systems and Signal Processing. 2007; 21(1): 108–124. doi: 10.1016/j.ymssp.2005.12.002
[7]Smith WA, Randall RB. Rolling element bearing diagnostics using the Case Western Reserve University data: A benchmark study. Mechanical Systems and Signal Processing. 2015; 64–65: 100–131. doi: 10.1016/j.ymssp.2015.04.021
[8]Lessmeier C, Kimotho JK, Zimmer D, et al. Condition Monitoring of Bearing Damage in Electromechanical Drive Systems by Using Motor Current Signals of Electric Motors: A Benchmark Data Set for Data-Driven Classification. PHM Society European Conference. 2016; 3(1). doi: 10.36001/phme.2016.v3i1.1577
[9]Wang B, Lei Y, Li N, et al. A Hybrid Prognostics Approach for Estimating Remaining Useful Life of Rolling Element Bearings. IEEE Transactions on Reliability. 2020; 69(1): 401–412. doi: 10.1109/TR.2018.2882682
[10]Lei Y, Han T, Wang B, et al. XJTU-SY Rolling Element Bearing Accelerated Life Test Datasets: A Tutorial. Journal of Mechanical Engineering. 2019; 55(16): 1–6. doi: 10.3901/JME.2019.16.001 (in Chinese)
[11]Chen L, Xu G, Tao T, et al. Deep Residual Network for Identifying Bearing Fault Location and Fault Severity Concurrently. IEEE Access. 2020; 8: 168026–168035. doi: 10.1109/ACCESS.2020.3023970
[12]Kumbhar SG, Sudhagar PE. An integrated approach of Adaptive Neuro-Fuzzy Inference System and dimension theory for diagnosis of rolling element bearing. Measurement. 2020; 166: 108266. doi: 10.1016/j.measurement.2020.108266
[13]Lin CJ, Jhang JY. Bearing Fault Diagnosis Using a Grad-CAM-Based Convolutional Neuro-Fuzzy Network. Mathematics. 2021; 9(13): 1502. doi: 10.3390/math9131502
[14]Lee S, Yu H, Yang H, et al. A Study on Deep Learning Application of Vibration Data and Visualization of Defects for Predictive Maintenance of Gravity Acceleration Equipment. Applied Sciences. 2021; 11(4): 1564. doi: 10.3390/app11041564
[15]Romanssini M, De Aguirre PCC, Compassi-Severo L, et al. A Review on Vibration Monitoring Techniques for Predictive Maintenance of Rotating Machinery. Eng. 2023; 4(3): 1797–1817. doi: 10.3390/eng4030102
[16]Lu J, Wang K, Chen C, et al. A Deep Learning Method for Rolling Bearing Fault Diagnosis Based on Attention Mechanism and Graham Angle Field. Sensors. 2023; 23(12): 5487. doi: 10.3390/s23125487
[17]Yoo Y, Jo H, Ban SW. Lite and Efficient Deep Learning Model for Bearing Fault Diagnosis Using the CWRU Dataset. Sensors. 2023; 23(6): 3157. doi: 10.3390/s23063157
[18]Wang X, Wang C, Zhu K, et al. A Mechanical Equipment Fault Diagnosis Model Based on TSK Fuzzy Broad Learning System. Symmetry. 2022; 15(1): 83. doi: 10.3390/sym15010083
[19]Tang Z, Hou X, Huang X, et al. Domain Adaptation for Bearing Fault Diagnosis Based on SimAM and Adaptive Weighting Strategy. Sensors. 2024; 24(13): 4251. doi: 10.3390/s24134251
[20]Xiao Z, Li D, Yang C, et al. Fault Diagnosis Method of Special Vehicle Bearing Based on Multi-Scale Feature Fusion and Transfer Adversarial Learning. Sensors. 2024; 24(16): 5181. doi: 10.3390/s24165181
[21]Ren J, Wen J, Zhao Z, et al. Uncertainty-Aware Deep Learning: A Promising Tool for Trustworthy Fault Diagnosis. IEEE/CAA Journal of Automatica Sinica. 2024; 11(6): 1317–1330. doi: 10.1109/JAS.2024.124290
[22]Das L, Gjorgiev B, Sansavini G. Uncertainty-aware deep learning for monitoring and fault diagnosis from synthetic data. Reliability Engineering & System Safety. 2024; 251: 110386. doi: 10.1016/j.ress.2024.110386
[23]Martins A, Fonseca I, Farinha JT, et al. Prediction maintenance based on vibration analysis and deep learning—A case study of a drying press supported on a Hidden Markov Model. Applied Soft Computing. 2024; 163: 111885. doi: 10.1016/j.asoc.2024.111885
[24]Zhong J, Lin C, Gao Y, et al. Fault diagnosis of rolling bearings under variable conditions based on unsupervised domain adaptation method. Mechanical Systems and Signal Processing. 2024; 215: 111430. doi: 10.1016/j.ymssp.2024.111430
[25]Mali Y, Upadhyay T. Fraud Detection in Online Content Mining Relies on the Random Forest Algorithm. SciWaveBulletin. 2023; 1(3): 13–20. doi: 10.61925/SWB.2023.1302
[26]Okoi Michael O. Natural Language Processing relevance to Online Business. SciWaveBulletin. 2023; 1(3): 37–42. doi: 10.61925/SWB.2023.1305
[27]Vartika V, William P. Machine Learning and Cloud Computing Case Study for Online App Usage Tracking. SciWaveBulletin. 2023; 1(4): 32–37. doi: 10.61925/SWB.2023.1405
[28]Yogeesh N. Fuzzy logic-based beat tracking in music signals. Musik in Bayern. 2023; 88(9): 145–157.
[29]Yogeesh N. Classroom leadership: An approach to educational psychology. International Journal of Early Childhood Special Education. 2022; 14(3): 3688–3691.
