Machine Learning and Vibration-Based Method for Anti-Friction Bearing Fault Severity Estimation

Incipient fault diagnosis of rolling element bearings using vibration, sound and tribological analysis: A literature review

Rolling element bearings are used in rotating machines in aviation, chemical, and nuclear industries. A failure to detect faults in the rolling bearing causes unexpected breakdown of rotating machines. Detecting bearing defects early on is still a challenge since micro- faults have less energy. Early-stage defects from fatigue, misalignment, overload and poor lubrication create low-energy signals that can be masked. This article summarises some key measurement methods, i.e. vibration and acoustic signal measurement, tribological parameter analysis, wear debris analysis, and thermal measurement, in rolling element bearings. All the methods will be evaluated with respect to a sensing mechanism, signal processing algorithm, micro-defect sensitivity, and operational constraints of the anti-friction bearing. To deal with the drawbacks of single sensors, we demonstrate the use of recent advancements in multi-sensor data fusion and machine learning models through examples where kinetic, acoustic and chemical metrics are fused for improved classification outcomes. In summary, multi-sensor data fusion greatly improves the accuracy of early fault detection, and it can greatly enhance predictive maintenance systems. Keywords: Vibration, Acoustic, Wear Debris, Tribology, Condition Monitoring, Bearing

Recent trends on bearing faults diagnosis of IM using vibration signal: a critical review

Evaluation of Vibration Characteristics Signatures for Condition Monitoring of Rotor Bearing System by Using Box-Behnken Experimental Design

To address the complexity of vibration in ball bearings with composite defects during actual operation, bearings play an imperative role in ensuring the smooth, low friction operation of rotating machinery by reducing friction. Bearings operating within the low to medium speed range are commonly used in a wide variety of industrial, automotive, and machinery applications. This paper proposes a dynamic model for the complex, combined effects of load, localized defects, and misalignment on the vibration characteristics of rotor-bearing systems. To establish a robust theoretical framework, dimensional analysis using the matrix method was first employed to identify the principal dimensionless groups governing the system's dynamic behaviour. Finally, the model proposed is verified by experiments, and the influence of misalignment, speeds, defect sizes, and loads on the vibration characteristics of the bearing is investigated. Response Surface Methodology (RSM), specifically a Box-Behnken Design (BBD), was implemented to systematically plan the experiments and quantify the interaction effects between the input variables. Vibration amplitudes were recorded under various combinations of applied load, defect severity, and misalignment levels. The BBD analysis provides a comprehensive model of the system's response, confirming that increases in both load and misalignment significantly elevate vibration amplitudes. The work of this paper can improve the accuracy of rotor compound fault identification and provide references for vibration control of rotating machines, fault diagnosis, and life prediction.

Mode entropy knowledge machine: a fully automated bearing fault diagnosis model for complex operating conditions

Accurate diagnosis of rolling bearing faults is critical to the reliability of industrial equipment. However, rolling bearings often operate under complex operating conditions, and with data imbalances and noise interference, fault diagnosis of them remains extremely challenging. To address these issues, a novel mode entropy knowledge machine (MEKM) framework for robust bearing fault diagnosis is proposed in this study. For MEKM, the mode entropy space is firstly constructed to decompose the vibration signal into intrinsic mode components, and the noise-resistant feature extraction and dimensionality reduction are realized by principal component analysis. Secondly, a fast classifier based on extreme learning machines is introduced, and its parameters are automatically adjusted through a particle swarm optimization to establish an adaptive extreme learning machine diagnosis model, ensuring optimal generalization under different load and speed levels. Then, a collaborative optimization paradigm is developed to coordinate mode entropy features and classifier parameters through fully automated learning, in which entropy-driven feature characterization guides the iterative refinement of decision boundaries, while classifier feedback dynamically improves the selectivity of entropy features. Finally, validation is performed on bearings with multiple operating conditions, and the results indicated that the MEKM outperformed conventional deep learning methods in terms of diagnostic accuracy and generalization ability. The work provides a theoretical basis and an industrially feasible solution for health monitoring of mechanical equipment.

Bearing fault classification in induction motors using current spectrum analysis and artificial neural network

Introduction. Bearing faults in induction motors are one of the primary causes of performance degradation and unexpected failures in industrial systems. Early fault detection remains challenging because conventional protection systems generally respond only after severe damage occurs. In addition, motor current signals exhibit nonlinear and complex characteristics, requiring advanced analysis techniques for accurate fault identification. Problem. Existing fault diagnosis methods often suffer from limited classification accuracy, dependency on specific operating conditions, and insufficient integration between spectral feature extraction and adaptive classification techniques. Goal. To develop a non-invasive bearing fault classification method based on current spectrum analysis and artificial neural network (ANN) for induction motor condition monitoring. Methodology. The proposed method utilizes fast Fourier transform (FFT) to transform motor current signals from the time domain into the frequency domain for spectral feature extraction. The extracted features are then processed using principal component analysis (PCA) for dimensionality reduction before being used as inputs to the ANN classifier. Experimental testing is conducted under 3 bearing conditions, namely normal, 7-ball fault, and 6-ball fault conditions, using 50 datasets for each condition. Results. The results demonstrate that the proposed method successfully identifies bearing conditions with high classification accuracy and strong separation characteristics in the PCA space. FFT analysis also reveals consistent spectral changes corresponding to fault severity, particularly in sideband components and energy distribution patterns. Scientific novelty. This work integrates FFT-based current spectrum analysis, PCA-based feature reduction, and ANN classification into a unified non-invasive diagnosis framework for bearing fault detection. Practical value. The proposed approach provides a simple, adaptive, and reliable solution for early bearing fault detection without requiring additional mechanical sensors, making it suitable for industrial condition monitoring applications. References 32, tables 4, figures 7.

Condition-based maintenance threshold determination for gearbox fault progression using vibration envelope features and accelerated life testing

Gearbox failures represent a critical problem faced by industrial machines, considering the impacts of such failures on machine reliability, efficiency, and maintenance cost. Despite the well-established use of vibration-based condition monitoring techniques for diagnosing the presence of faults, few attempts have been made in transforming information about fault progression into thresholds that could be used in making condition-based maintenance (CBM) decisions. In this work, a CBM system for analysing gearbox fault progression based on vibration envelope features is presented, alongside accelerated life testing. An experimental approach has been adopted, whereby accelerated life testing was performed to induce gearbox degradation progressively. Then, vibration data were analysed through the envelope technique, out of which six vibration envelope features, namely RMS Envelope, Kurtosis, Crest Factor, Peak Amplitude, Envelope Energy, and Sideband Energy Ratio, were derived and analysed based on their sensitivity through correlation, monotonicity, trendability, and separability tests. Thereafter, a composite health index based on the most sensitive features was formulated, and a multilevel maintenance threshold system consisting of Alert, Warning, and Critical levels was created. The findings show that Envelope Energy, Sideband Energy Ratio, and Kurtosis have the highest sensitivity to degradation and can accurately represent the evolution of gearbox faults. The composite health index shows a strong correlation with the extent of degradation (R2 = 0.962) and successfully discriminates between different health states of the gearbox. The developed framework for determining maintenance thresholds achieves an accuracy of 94.9%, which allows accurate identification of maintenance intervention phases.