Jul 2026· Applied and Computational Engineering· Vol 250, pp. 22-31· 0 citations
TL;DR
Experimental results show that MFF-Net achieves stable training, high classification accuracy, and clear feature separation, offering a feasible solution for lightweight online fault diagnosis in power electronic systems.
Abstract
Three-phase inverters are widely used in renewable energy conversion, industrial drives, and energy storage systems. Their IGBTs and other power switching devices often work for long periods under high-frequency and high-current conditions, which makes fault diagnosis an important issue for reliable operation. Traditional diagnosis methods usually depend on manually designed features obtained from Fourier transform, wavelet analysis, or related signal-processing tools. Although these features are interpretable, their performance is closely tied to expert experience. Purely data-driven deep learning models can learn features from raw waveforms, but they often show limited physical interpretability and may overfit when fault samples are insufficient. This paper proposes a lightweight Multi-view Fault Feature Fusion Network (MFF-Net) for fault waveform classification in three-phase inverter circuits. The model contains a lightweight one-dimensional convolutional neural network branch for temporal waveform representation and a physics-guided branch for extracting single-phase statistics, three-phase imbalance indices, zero-sequence components, and multi-band harmonic energy ratios. A gated fusion module is then used to combine the two feature groups according to sample-specific fault characteristics. A simulated dataset with 3,200 samples is built under normal, overcurrent, phase-loss, and IGBT bridge-arm open-circuit conditions. Experimental results show that MFF-Net achieves stable training, high classification accuracy, and clear feature separation, offering a feasible solution for lightweight online fault diagnosis in power electronic systems.
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 progressive three-stage time–frequency learning framework that identifies series arc faults directly from normalized current waveforms and provides robust discrimination across unseen measurement sessions within the evaluated load categories and operating conditions is presented.
Seoyoung Jeon, Won-Kyu Choi, Sungsoo Kwon et al.· Italian National Conference...· 0 citations
In renewable-powered distribution systems and microgrids, reliable cable condition monitoring is essential for operational security and early fault detection. Sheath circulating current signals provide valuable information for identifying grounding abnormalities and incipient faults, but their diagnosis is difficult because the signals exhibit strong inter-phase coupling, and fault samples are limited in practice. This study proposes a dual-view Mixup-ResNet framework for fault diagnosis of cable sheath circulating current signals. Specifically, physical-range normalization is employed to retain magnitude-related fault information and inter-phase proportional relationships, while sample-wise z-score normalization is used to emphasize waveform morphology. These two complementary views are concatenated to form a six-channel input for a lightweight one-dimensional residual network, which is trained with Mixup, label smoothing, dropout, and cosine annealing. On an ATP-EMTP-generated eight-class dataset, the proposed method achieves an average accuracy of 91.50%, a weighted F1-score of 91.69%, and a macro F1-score of 91.69% under a unified 5 × 5 repeated stratified cross-validation protocol. Additional tests under 12% relative Gaussian noise show that the method maintains competitive Gaussian-noise tolerance within the tested condition, although broader field disturbances remain to be further validated. These findings suggest that the proposed method has potential for small-sample fault diagnosis of cable sheath circulating current signals and provides a preliminary basis for intelligent cable condition monitoring.
Haiqi Yang, Jinwei Mao, Bo Zhang et al.· Energies· 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
A hybrid two-stage machine learning pipeline that decouples detection from classification is proposed, and the direction of the zero-sequence signature is found to be system-dependent, motivating a learned decision boundary in place of a fixed relay threshold.
Sahil Manikshete, A. Gujarathi, Thanh Long Vu et al.· 0 citations
Power-switch open-circuit (OC) faults distort electric-vehicle (EV) inverter phase currents and require fast on-board diagnosis for fault-tolerant control. Trajectory-image methods encode the α–β current-vector trajectory as a binary image and classify it with a convolutional neural network (CNN); however, the baseline uses 6.46×105 parameters and 3.31×107 multiply–accumulate (MAC) operations per inference, which is costly for motor-control microcontrollers (MCUs). Here, each one-cycle trajectory is represented by a 36-dimensional normalized angular sector-energy vector and classified by a compact two-stage multilayer perceptron. Sector accumulation averages zero-mean measurement noise in the representation, without relying on noise-augmented training. The locating stage uses 1.58×104 parameters and 1.56×104 MACs per inference, 97.55% and 99.95% fewer than the baseline CNN; the complete pipeline runs on a TI F28379D in 0.52 ms. On measured resistive-load currents, both methods reach 100% accuracy from 40 to 20 dB, whereas the proposed method remains more accurate at 15 and 10 dB, including under 88% phase-current unbalance. A supplementary balanced RL-load experiment preserves 100% clean accuracy, confirming MCU-executable diagnosis under a lagging power-factor load for embedded EV inverter protection.
Mingxing Fang, Wenxu Yan, Wenyuan Wang· World Electric Vehicle Journ...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
