TL;DR
Results show that using statistical vibration features with ensemble classifiers is a good way to diagnose multi-class bearing faults and establishes a comprehensive benchmark for ML- and DL-based rolling bearing FDD.
The findings affirm the efficacy of XGBoost in bearing fault classification and emphasise the diagnostic value of carefully selected time-domain features, as well as suggesting strong potential for deploying such models in real-time condition monitoring and predictive maintenance systems.
Accurate detection of faults within rotary machinery components is vital for assuring the structural reliability of equipment in manufacturing facilities and power generation systems. Automatic fault detection is an increasingly common approach whereby data acquired from sensors are analyzed by machine learning algorithms to distinguish between normal and faulty component states. Vibration signatures recorded from healthy and faulty ball bearings, taken from the Case Western Reserve University (CWRU) 12k Drive End database, were utilized in the present study. These vibration signatures were segmented into windows of 0.1 s duration. Standard statistical time-domain features (root mean square, kurtosis, peak value and crest factor) were extracted from each sample. A random forest (RF) classifier was trained on vibration data from healthy bearings and several types of faulty bearings, partitioned into training (70% of data) and testing (30%) sets. The RF classifier distinguished healthy from faulty bearings with 98.8% accuracy across ball, inner race, outer race and normal categories. Feature importance analysis indicated that RMS (50% importance) and peak value (28% importance) were the most influential contributors, together comprising about 78% of the decision making. The RF model further recorded a recall of 1.00 for the normal baseline category, indicating that healthy bearings were identified with zero false negatives. The results demonstrate that ensemble machine learning, combined with statistically significant time-domain features, provides a highly accurate and physically interpretable solution for industrial condition monitoring, serving as a foundation for more advanced rotor-dynamic fault detection.
It is confirmed that traditional machine learning models, optimized through manual feature engineering, can provide a ‘high-precision, low-risk’ solution for bearing fault diagnosis and offers significant reference value for the intelligent operation of aero-engines and other industrial equipment.
The near-perfect linear separability indicates that the dataset’s binary, controlled-laboratory labelling rather than intrinsic bearing-degradation physics drives the clean classification, and validation on 500–1000 or more samples with progressive-degradation labelling is essential before any operational claim can be supported.
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.
A new predictive maintenance model that incorporates multi-domain feature extraction, and hybrid feature selection approach and optimized ensemble classifier to facilitate high-fidelity fault detection in four operational modes is proposed, affirming its suitability for deployment in industrial cyber-physical monitoring systems.