Jul 2026· European Conference on Artificial Intelligence· pp. 1-7· 0 citations· 16 references
Abstract
Squirrel Cage Induction Generators are widely used in renewable energy systems, but their reliability is threatened by stator short-circuit faults. Traditional steady-state diagnostic methods are often rendered ineffective in modern applications because closed-loop controllers actively mask fault signatures. To address this challenge, this work introduces an event-driven fault classification framework specifically designed to detect and classify masked stator faults, namely inter-turn and inter-winding short-circuits. Instead of computationally expensive continuous signal analysis, the proposed methodology isolates brief, high-frequency transient disturbances using an adaptive, derivative-enhanced event detection mechanism. Discriminative features are extracted from these localized transient intervals, with a focus on derivative-based voltage representations. Finally, a novel confidence-weighted aggregation strategy is introduced to combine event-level predictions into robust file-level decisions. Experimental validation using a high-resolution dataset demonstrates that the proposed framework significantly improves diagnostic accuracy. The results show that derivative-based voltage features provide superior class separability, achieving a file-level Area Under the Curve of approximately 0.89, demonstrating the effectiveness of the targeted transient evaluation strategy in closed-loop systems.
Inter-turn short-circuit faults are a common failure mode in permanent magnet synchronous motors, and their early diagnosis is essential for ensuring safe and reliable operation. In practical applications, however, training data are often contaminated with noisy labels, while incipient fault features are weak and exhibit limited inter-class separability, which further aggravates label misassignment and mixing. To enhance data usability and label reliability, this paper proposes a diagnostic framework integrating high-frequency signal injection with unsupervised learning. High-frequency signal injection is used to strengthen fault-related current features, followed by autoencoder-based feature extraction, UMAP dimensionality reduction, and HDBSCAN clustering to group samples, after which a majority voting strategy is applied for sample relabeling. Experimental results demonstrate that the proposed method can effectively improve noisy-label correction performance even under high noise ratios and low fault severity, thereby providing reliable data for subsequent fault diagnosis.
Haoguang Wang, Baoye Song, Wei Xin et al.· International Conference on...· 0 citations
Rotor inter-turn short-circuit (ITSC) faults in synchronous generators introduce electromagnetic asymmetries that can lead to torque ripple, unbalanced magnetic pull, and progressive mechanical degradation. While most existing studies focus on binary classification and severe fault conditions, the assessment of incipient rotor ITSC severity using displacement-sensitive vibration measurements remains relatively underexplored. This paper proposes a vibration-based diagnostic framework for multi-class severity classification of rotor ITSC by integrating an eddy-current displacement sensor with physically motivated feature extraction. An 18-dimensional hybrid feature set is designed to characterize electromechanical modulations induced by rotor electromagnetic asymmetry. Using an XGBoost classifier with leave-one-out cross-validation, the proposed approach achieved 90.56% overall accuracy, including 99% recall for healthy operation and 87% recall for mild fault conditions. The results suggest that displacement-sensitive vibration analysis enables effective severity-aware diagnosis of rotor ITSC with minimal sensor requirements.
O. Rozhon, Serge Pacome Bosson, Š. Janouš et al.· 0 citations
Permanent magnet synchronous motors (PMSMs) are broadly used in diverse applications due to their inherent advantages. Open-circuit faults (OCFs) are among the major fault classifications in PMSMs, posing significant concerns due to their contribution to torque ripples, vibrations, and efficiency degradation. Therefore, accurate and real-time OCF diagnosis is essential for reliable operation and predictive maintenance practices. This underscores the importance of a robust diagnostic framework that enables early fault detection and localization, supports embedded integration, and requires no additional dedicated sensors. However, existing studies rarely address these requirements together. To overcome these limitations, this article proposes a novel OCF diagnostic framework that fuses features derived from multiple strategies, including wavelet energy-based features, frequency-domain features extracted from current waveforms, and speed measurement data. The extracted feature vector is used as input to a lightweight deep neural network. The proposed approach enhances interpretability and enables seamless embedded integration compared to conventional raw-data-driven machine learning models. In addition, an extended refinement layer is incorporated to enable integrated fault detection and classification for OCF while enhancing diagnostic transparency. The effectiveness of the proposed method is demonstrated through MATLAB/Simulink simulations using the PLECS Blockset and further validated in real-time with an RTBox-based hardware-in-the-loop setup using a C2000 launchpad. Furthermore, experimental validation is conducted using a domain-adaptation strategy based on transfer learning. Performance evaluation confirms diagnostic accuracy exceeding 99% across varying operating conditions. The validation process achieves fault detection within 22% of a fundamental electrical cycle, with fault localization occurring within 40% of an average, demonstrating the robustness and adaptability of the proposed method. A sensitivity analysis of the proposed algorithm’s feature vector validates the effectiveness of high-frequency features. Furthermore, the risk distribution matrix provides insights supporting informed maintenance decisions.
Nimesh Jayasena, Battur Batkhishig, B. Nahid-Mobarakeh et al.· IEEE Open Journal of Industr...· 0 citations
A novel perspective on noninvasive diagnostics by integrating advanced signal processing with deep learning classifiers is offered, indicating that appropriate signal preprocessing enhances feature representation quality, indicating that the choice of transform method directly impacts diagnostic accuracy.
Konrad Górny, Wojciech Pietrowski· Compel· 0 citations
Inter-turn short circuits (ITSCs) in induction motor (IM) windings are among the most critical and frequent faults in industrial environments, as they can rapidly evolve into severe damage, leading to unplanned downtime and costly maintenance. To enhance the reliability of IMs, this paper proposes a machine learning–based diagnosis method dedicated to ITSC failures. The developed diagnostic tool combines a support vector machine (SVM) classifier with Fisher’s ratio (FR)-based feature selection. The proposed framework uses experimentally acquired current signals under healthy conditions and five ITSC fault severity levels (1%–5%), evaluated across four load conditions (25%, 50%, 75%, and 100%). Each signal is segmented into 200 non-overlapping segments (500 samples each), from which nine time-domain features are extracted to capture fault-related characteristics. These features are then used for training and testing a SVM classifier capable of distinguishing between healthy states and levels of severity of ITSC faults. To optimize the classification process, the Fisher’s ratio (FR) algorithm is employed to select the most informative features while discarding those with low relevance. Our findings unveiled that the proposed hybrid FR-SVM-based diagnosis achieves high diagnostic accuracy ranging from 99.54% to 100%. Furthermore, the outcomes prove that the integrated technical framework (time-domain features + FR + SVM) provides zero false alarms and a balanced diagnostic system that combines computational speed with high precision.
H. Zaimen, T. Thelaidjia, Makhlouf Chouki et al.· International Journal of Ele...· 0 citations
Fault diagnosis in power electronic systems is challenging when fault categories geometrically overlap within the measurement space, limiting class separability and introducing classification ambiguity. This work proposes an edge-oriented hybrid fault-space restructuring methodology that utilizes UMAP-based embeddings and hierarchical clustering to group overlapping fault conditions into robust hybrid representations. Subsequently, supervised machine learning models execute the final classification over this optimized space. Validation was conducted using a large-scale synthetic dataset generated via real-time hardware-in-the-loop (HIL) simulation, evaluating electrical measurements from three-dimensional RMS values to 60-dimensional instantaneous waveforms. Tested with Decision Tree and Random Forest algorithms, the restructuring strategy significantly improves robustness under geometric ambiguity compared to conventional classification without space restructuring. Specifically, low-dimensional measurements achieved F1-score improvements of approximately 72% and 46% for the Decision Tree and Random Forest algorithms, respectively, while high-dimensional measurement configurations still exhibited significant improvements of 36% and 52%. Consequently, these results confirm that the combined restructuring and classification pipeline is highly effective across the analyzed measurement dimensionalities, establishing a dependable cluster-based diagnostic strategy that enhances classification robustness while accepting a trade-off in individual fault-isolation granularity. Finally, hardware deployment experiments on a Raspberry Pi 4 platform demonstrated the feasibility of executing the trained classifiers for real-time inference under constrained computational environments. The experimental evaluation validated real-time execution capabilities, achieving sub-millisecond inference latencies (as low as 0.32 ms), a memory footprint under 0.14 MB, and processing rates exceeding 2600 inferences per second using lightweight Decision Tree classifiers. Ultimately, these findings indicate that the proposed strategy improves fault detection across the evaluated measurement configurations while ensuring a highly viable execution on resource-constrained devices once the classifiers are trained.
J. M. García-Campos, A. M. Alcaide, A. Letrado-Castellanos et al.· Electronics· 0 citations