Fault identification of vacuum on-load tap changers (OLTCs) is essential for reliable transformer operation and condition-based maintenance. Existing diagnostic methods often rely on single-modality feature settings or treat feature selection and SVM hyperparameter optimization as separate steps. These practices may limit the use of complementary fault information and lead to suboptimal model configurations. To address these limitations, a multimodal PKO-SVM framework is proposed by integrating vibration-acoustic feature-level fusion with joint feature selection and SVM hyperparameter optimization. A dataset containing 250 samples from five operating conditions was constructed. Vibration features were extracted to characterize mechanical impacts and transient dynamic responses, whereas acoustic features were derived to describe spectral-envelope and time-frequency texture variations during switching. In the proposed framework, PKO is used to jointly optimize the feature-selection mask and the SVM hyperparameters C and γ, thereby reducing empirical parameter dependence and balancing recognition performance and feature compactness through a unified fitness function based on mean Macro-F1 and a feature-count penalty term. Repeated experiments with an independent test set were conducted for evaluation. The PKO-SVM model using the fused vibration-acoustic feature set achieved an accuracy of 92.53% ± 2.98% and a Macro-F1 of 0.9252 ± 0.0299. Compared with acoustic-feature-only and vibration-feature-only inputs, the fused vibration-acoustic feature set improved the average accuracy by 8.43 and 2.96% points, respectively. Comparisons with MLP/BPNN, TabNet, Random Forest, GA-SVM, and XGBoost, together with feature-contribution analysis, further supported the effectiveness and interpretability of the proposed multimodal PKO-SVM framework. These results indicate that the multimodal PKO-SVM framework has the potential to provide a reliable and compact diagnostic model for online condition assessment of vacuum OLTCs.
To address the challenges of conventional feature extraction methods in capturing nonlinear dependencies within fault signals and reducing feature redundancy during hydropower units fault diagnosis, this paper proposes a feature selection framework integrating the minimum redundancy maximum relevance (mRMR) criterion with copula entropy (mRMR-CE). This framework utilizes CE to capture both linear and nonlinear dependencies in vibration signals. Combined with the mRMR criterion to suppress feature redundancy, it achieves stable selection of highly discriminative features. In noisy environments and with various classifiers, the method shows strong performance and stability. To validate the effectiveness of the proposed frame, seven feature selection approaches—CE, mRMR, mRMR-CE, Pearson, Principal Component Analysis, Hibert-Schmidt independence criterion—Lasso, and concrete autoencoder —were applied to the training samples during the feature selection stage. Lastly, the chosen features were input into four different types of classifiers for training and testing: support vector machine, Random Forest, multi-layer perceptron, and extreme gradient boosting. Experimental results demonstrate that the proposed method exhibits outstanding performance on both the Case Western Reserve University (CWRU) bearing fault dataset and the Unit 3 dataset from a hydropower plant. On the CWRU dataset, it achieved an average precision of 99.76% and an average F1 score of 96.57%, while on the Unit 3 dataset, it attained a 98.96% average accuracy and an average F1 score of 98.97%. These results significantly outperform traditional feature selection methods while demonstrating high stability and robustness.
Bo Li, Jiahao Li, Guangtao Zhang et al.· Engineering Research Express· 0 citations
To address the issues of parameter dependency on empirical settings, insufficient fault feature extraction capability, and limited classification accuracy in rolling bearing fault diagnosis using Variational Mode Decomposition (VMD), a novel fault diagnosis method based on adaptive signal decomposition and intelligent classification integration is proposed. The VMD parameters are adaptively optimized using the Subtraction-Average-Based Optimizer (SABO), and a kurtosis–correlation criterion is introduced to select a single fault-sensitive intrinsic mode function, from which time-domain features are extracted to construct fault feature vectors. The Moth-Flame Optimization Algorithm (MFOA) is employed to optimize the parameters of the Kernel Extreme Learning Machine (KELM) for fault state identification. From the perspective of methodological symmetry, the averaged population update of SABO is invariant to the ordering of search agents, VMD exhibits equivalence under permutation of mode labels, and KELM constructs the sample similarity matrix using a symmetric kernel function. These symmetry-related structures are integrated into the parameter optimization, modal decomposition, and fault classification stages of the proposed method. Experimental validation using the CWRU rolling bearing dataset demonstrates that the proposed method reaches a fault recognition accuracy of 96.73%, outperforming other comparative models and exhibiting superior diagnostic precision and robustness.
Reliable fault diagnosis in rotating machinery is challenging due to the nonlinear and non-stationary nature of vibration signals. Although time–frequency analysis is widely used, it cannot capture the cross-scale coupling between amplitude-modulated (AM) and frequency-modulated (FM) components that carry essential diagnostic information. This study applies Holo-Hilbert Spectrum Analysis (HHSA) to extract amplitude–frequency modulation features and integrates them with six machine learning classifiers to identify four fault conditions. Random Forest, K-Nearest Neighbors, and Logistic Regression achieve accuracies of up to 99.95%, yielding higher accuracy than Fast Fourier Transform-based features. The proposed framework employs an HHSA-based feature extraction pipeline that effectively captures AM–FM coupling in nonlinear vibration signals. It also provides higher discriminative capability than traditional spectral approaches and maintains robustness across multiple classifiers. This method offers high diagnostic accuracy and strong potential for industrial predictive maintenance. Future work will focus on improving computational efficiency and evaluating the framework under more diverse and realistic operating conditions.
Received: 10 September 2025 | Revised: 20 April 2026 | Accepted: 10 June 2026
Conflicts of Interest
The authors declare that they have no conflicts of interest to this work.
Data Availability Statement
The VBL-VA001 datasets that support the findings of this study are openly available at https://doi.org/10.1007/s42417-023-00959-9, reference number [44].
Author Contribution Statement
Van-Trung Nguyen: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Writing - original draft, Writing - review & editing, Visualization. Ba-Tan Le: Investigation, Data curation, Writing - original draft, Writing - review & editing. Van-Phuong Dao: Methodology, Validation, Writing - review & editing.
Van-Trung Nguyen, Ba-Tan Le, van-Phuong Dao· Journal of Computational and...· 0 citations
This paper presents a systematic framework for real-time fault-type identification in rolling element bearings using vibration signal analysis. The Case Western Reserve University bearing dataset is employed as the primary data source, comprising vibration signals recorded under different load conditions (0–3 HP). Initially, Envelope Analysis (EA) is applied to extract fault-related characteristic frequencies. While computationally efficient, EA successfully identifies fault features in 66.67% of the signals but shows limitations under noisy and spectrally smeared conditions. To address this, Cepstrum Prewhitening Analysis (CPA) is selectively applied to unresolved signals, achieving a 75% detection success in these cases and improving the overall detection rate to 91.67%. Thereafter, 36 time-domain, frequency-domain, and EA-CPA based features were extracted from segmented vibration signals. A sequential feature optimization strategy comprising variance threshold filtering, correlation analysis, Z-score normalization, ANOVA F-test feature ranking, and Recursive Feature Elimination reduced the feature set to the 10 most discriminative features. To eliminate sample- and group-level information leakage, a leakage-free Nested GroupKFold framework was developed, in which preprocessing, feature selection, and hyperparameter optimization was performed exclusively within the training folds using GridSearchCV. Six machine learning classifiers, namely Random Forest, XGBoost, LightGBM, Support Vector Machine, K-Nearest Neighbors, and Logistic Regression, were comparatively evaluated. XGBoost achieved the highest mean classification accuracy of 97.62%, while RF attained a comparable accuracy of 97.51% with lower fold-to-fold variation, indicating superior robustness and stability. Consequently, RF was selected for independent cross-condition validation, in which it was trained solely on the 0 HP operating condition and evaluated on the unseen 1 HP, 2 HP, and 3 HP datasets, demonstrating strong generalization across varying load conditions. Feature importance analysis further confirmed that the characteristic bearing defect frequencies (BPFO, BPFI, and BSF) are the dominant contributors to classification performance. The proposed framework provides an effective balance between diagnostic accuracy, robustness, interpretability, and computational efficiency, making it a promising solution for intelligent condition monitoring and predictive maintenance of rotating machinery.
Rohit Mishra· Journal of engineering and a...· 0 citations
To address the issues of insufficient feature extraction and low localization accuracy in distribution network fault diagnosis, this study proposes a fault classification and localization method based on APC-SVM and PC-AZOA. The model performs a simultaneous decomposition of three-phase signals using multivariate variational modal decomposition and employs the energy entropy of each model component as the feature vector; During the classification stage, the method integrates electrical and physical constraints, introducing three-phase energy imbalance and variance into the support vector machine ’ s parameter optimization process for the first time to dynamically adjust the penalty factor and kernel parameters; finally, a traveling wave propagation time error model is constructed, and an adaptive zebra optimization algorithm constrained by physical information is proposed. By innovatively embedding prior physical knowledge into the search space constraints, the method effectively suppresses invalid searches and improves convergence efficiency. Experimental results show that the model achieves a classification accuracy of up to 98.4% with a positioning error below 1%, demonstrating both high precision and high efficiency.
Dahua Li, Xinrui Yang, Yu Song et al.· 2026 IEEE International Conf...· 0 citations
The effective and dependable functioning of high-speed permanent-magnet brushless DC motors used in aerospace and industry relies on motor fault classification and optimisation of efficiency. Accurate problem detection and diagnosis are critical for preserving system stability and performance, while attaining entirely fault-free devices is impossible according to dependability theory. This research proposes a state-of-the-art hybrid framework for motor fault classification that makes use of mutual information from current signals to effectively extract features. The representation is built on top of statistical characteristics, and to uncover hidden patterns in the data, deep features are retrieved using an adaptively trained DNN employing t-SNE visualisation. Afterwards, the Extreme Gradient Boosting (XGBoost) technique is used to integrate and classify these features. Particle Swarm Optimisation (PSO) is then used to automatically tweak the model parameters and improve performance. The results show that the suggested PSO-XGB-DNN model improves diagnostic accuracy by surpassing traditional methods, with a high classification accuracy of 97.15 percent. Finally, motor fault categorisation is made much more efficient, reliable, and operationally efficient by combining statistical and deep learning algorithms. This also improves predictive maintenance capabilities.
B. M. Reddy, G. Meghana, R. N. Sri et al.· 2026 7th International Confe...· 0 citations
