A reliable rolling bearing fault diagnosis method based on Titan

Comparative Evaluation of Machine Learning Algorithms for Fault Diagnosis in Automotive Press Lines

The results show that the proposed Condition Monitoring (CM) approach significantly reduces resource waste and prevents costly downtime, offering a practical and scalable asset management model for industrial applications.

An Interpretable and Calibrated Energy-Based Model for Bearing Fault Diagnosis and Out-of-Distribution Detection

Predictive maintenance of rotating machinery in industrial settings requires bearing fault diagnosis that is both accurate and auditable by maintenance engineers. Methods that achieve high classification accuracy typically operate as closed-box deep learning models, while methods that provide interpretability rarely report probability calibration, out-of-distribution (OOD) detection, or pairwise statistical significance. This paper addresses this gap with an Energy-Based Model (EBM) trained via Stochastic Gradient Langevin Dynamics on physics-informed features, in which the energy gradient with respect to each input feature provides intrinsic interpretability without post-hoc surrogates. A single forward pass simultaneously yields classification logits, a calibrated probability output, and a free-energy score for OOD detection. On three benchmark datasets (CWRU, MFPT, Paderborn) under file-level splitting and across five training seeds, the method attains 93.24% accuracy on Paderborn with an Expected Calibration Error of 0.017, less than half that of a Random Forest baseline. The same energy score separates inputs drawn from datasets absent at training time and preserves accuracy under additive feature noise. A leave-one-operating-condition-out evaluation bounds the operating envelope, in which three of four unseen conditions transfer with moderate loss and the lowest-speed condition does not. The results indicate that classification, probability calibration, OOD detection, noise robustness, and feature-level interpretability can be jointly delivered by a single energy-based model on bearing fault diagnosis benchmarks, without auxiliary classifiers or post-hoc explanation modules.

Motor-Current-Based Bearing Fault Detection Under Unseen Operating Conditions

The results show that instability is broader than the TF–CNN pipeline but is not uniform across model families: TF candidates were clearly stronger for Targets A and C, feature-based models were stronger for Target B, and Target D remained mixed and protocol-sensitive.

Multi-Class Fault Detection and Diagnosis of Rolling Bearings: a Machine Learning Approach

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.

BEARING EARLY FAULT DETECTION AND REMAINING USEFUL LIFE ESTIMATION WITH KNOWLEDGE-INFORMED MACHINE LEARNING

Hierarchical Approach to Rolling Bearing Failure Detection Using Spectrogram Images Classification

Early detection of rolling bearing faults is essential for preventing failures in rotating machinery. Although vibration-based analysis is widely used for fault diagnosis, it requires additional sensors, such as accelerometers, which increases system cost and complexity. Motor current signals, inherently available in industrial drive systems, offer a cost-effective alternative; however, their indirect sensitivity to mechanical defects makes reliable fault diagnosis challenging. This paper proposes a hierarchical motor current-based diagnostic framework that integrates domain knowledge with data-driven image analysis. Raw current signals are transformed into time-frequency spectrograms, which provide enhanced representations of faults compared to pure time or frequency-domain features. The proposed approach consists of three stages. First, bearing health is characterized using a rule-based method to compute the characteristic frequencies associated with different fault types, incorporating domain knowledge into the diagnosis process. Second, this information is used to guide a hierarchical classification strategy that separates health assessment from fault localization. Finally, global image descriptors extracted from the spectrograms are employed to classify the bearing condition as healthy or faulty and to identify the fault location (internal or external). By explicitly combining physics-informed fault characterization with image-based machine learning, the proposed approach improves interpretability while maintaining competitive diagnostic performance. The effectiveness of the method is validated on a benchmark bearing dataset, demonstrating its potential for cost-effective industrial condition monitoring.