Fault Diagnosis Research Articles

Study of control methods for three-level, five-level and multilevel inverters

With the rapid development of power electronics technology, multilevel inverters have gained wide attention in high-voltage and high-power applications due to their advantages of high output waveform quality and low switching loss. The article firstly introduces the basic topologies and control strategies of three-level inverters, including midpoint clamp type, flying capacitor type and H-bridge cascade type, etc. It focuses on analyzing the principles and characteristics of the control methods, such as space-vector pulse-width modulation and selective harmonic elimination. The research progress of five-level inverters is summarized, and the advantages and disadvantages of different topologies and their control techniques, such as modular multilevel and hybrid cascade types, are discussed. Subsequently, the article outlines the development trend of higher level number inverters and discusses the key technical issues facing multilevel inverters, such as voltage balance control, fault diagnosis and fault tolerance. The future research directions of multilevel inverter control methods are envisioned, including the application of intelligent control algorithms and the development of high-efficiency and high-reliability control strategies. This review provides a theoretical foundation and reference for the in-depth research of multilevel inverter control technology.

CNN-Based Fault Classification in Induction Motors Using Feature Vector Images of Symmetrical Components

Motor Current Signature Analysis (MCSA) is a commonly used non-invasive method for diagnosing faults in electric motors. Although MCSA provides significant advantages—current signals are easy to acquire and inherently robust against noise—this study aims to further enhance its diagnostic capabilities by focusing on symmetrical components. Three-phase stator current signals are converted into zero, positive, and negative sequence components, and their time-domain feature vectors are systematically integrated into a single image representation. A Convolutional Neural Network (CNN) is then employed for fault classification. The proposed method is model-free, requiring no explicit motor model, which offers greater flexibility compared to model-based techniques. Validation experiments were conducted on a rotor kit test bench under seven different conditions (one healthy condition and six mechanical/electrical fault conditions), with fault severities chosen to reflect practical scenarios. The symmetrical components-based image classification method demonstrated superior performance, achieving 99.76% classification accuracy and outperforming a widely used Short-Time Fourier Transform (STFT)-based spectrogram approach. These findings highlight that integrating all symmetrical component information into one image effectively captures each fault’s distinct behavior, enabling reliable diagnostic outcomes. By leveraging the distinct variations in zero, positive, and negative components under fault conditions, the proposed method offers a powerful, accurate, and non-invasive framework for real-time motor fault diagnosis in industrial applications.

A Full-Time Cross-Business Interaction Method Based on Multi-Source Data

In order to effectively integrate and utilize power multi-source data, avoid information islands, and improve the comprehensive utilization of power data, a full-time and space-time cross-business interaction method based on multi-source data is studied. The intelligent sensing business layer obtains power-related data from various systems, such as geographic information systems (GIS) and distribution management systems (DMS). The full spatiotemporal data fusion business layer extracts spatiotemporal features of power data and matches spatiotemporal correlations between power data and networks. Construct a power spatiotemporal grid coding database, integrate power data from different sources, form a spatiotemporal data system, and store it in the all-spatial and temporal data storage business layer. The application business layer calls the required power data from the all-spatial and temporal data storage business layer to realize the status of power equipment prediction, remote fault diagnosis, intelligent operation and maintenance decision-making, and other functions, among which the power equipment status prediction module combines the differential autoregressive integrated moving average (ARIMA) model and the long short-term memory network (LSTM) model to achieve fusion prediction of power equipment status. Enterprise service bus (ESB), as a communication hub that realizes cross-business interaction of all-time and space-time data in electric power, ensures the smoothness of data interaction between businesses and eliminates the phenomenon of information “islands” between businesses. The experimental results show that this method can effectively fuse power spatio-temporal data, with a data fusion speed of 12 GB/s, capable of processing 513 data fusion tasks simultaneously, with a processing time of only 80 seconds. The CPU utilization rate is 60%, the memory occupation is 180 MB, the accuracy is 92%, the recall rate is 90%, and the F1 score is 91, providing strong support for the intelligent operation and maintenance of power systems.

Deep Learning-Based Fault Diagnosis via Multisensor-Aware Data for Incipient Inter-Turn Short Circuits (ITSC) in Wind Turbine Generators

Wind energy is a vital pillar of modern sustainable power generation, yet wind turbine generators remain vulnerable to incipient inter-turn short-circuit (ITSC) faults in their stator windings. These faults can cause fluctuations in the output voltage, frequency, and power of wind turbines, eventually leading to overheating, equipment damage, and rising maintenance costs if not detected early. Although significant progress has been made in condition monitoring, the current methods still fall short of the robustness required for early fault diagnosis in complex operational settings. To address this gap, this study presents a novel deep learning framework that involves traditional baseline machine-learning algorithms and advanced deep network architectures to diagnose seven distinct ITSC fault types using signals from current, vibration, and axial magnetic flux sensors. Our approach is rigorously evaluated using metrics such as confusion matrices, accuracy, recall, average precision (AP), mean average precision (mAP), hypothesis testing, and feature visualization. The experimental results demonstrate that deep learning models outperform machine learning algorithms in terms of precision and stability, achieving an mAP of 99.25% in fault identification, with three-phase current signals emerging as the most reliable indicator of generator faults compared to vibration and electromagnetic data. It is recommended to combine three-phase current sensors with deep learning frameworks for the precise identification of various types of incipient ITSC faults. This study offers a robust and efficient pipeline for condition monitoring and ITSC fault diagnosis, enabling the intelligent operation of wind turbines and maintenance of their operating states. Ultimately, it contributes to providing a practical way forward in enhancing turbine reliability and lifespan.

Multi-Condition Rolling Bearing Fault Denoising Method and Application Based on RIME-VMD

To improve the stability of rolling bearing fault signal denoising under different working conditions, this study proposes a multi-condition rolling bearing fault denoising method based on RIME-VMD. According to the characteristics of rolling bearing fault signal, RIME (frost and ice algorithm) is utilized to obtain adaptive optimization of the modal number and penalty factors in VMD (variational mode decomposition) algorithm, and then the optimized core parameters are input into VMD to decompose the rolling bearing fault signal. The fitness function is established by introducing the fusion of power spectrum entropy and kurtosis value; these intrinsic modal functions (IMFs) with high correlation based on original rolling bearing fault signal are selected to reconstruct the rolling bearing fault denoising signal. The experimental results show that the RIME-VMD method can effectively remove most of the noise in the rolling bearing fault signal, and the performance evaluation indexes of this method are better than other existing optimization algorithms. The research achievement of this study can provide effective data support for the fault diagnosis of bearing equipment.

TH-RotatE: A Hybrid Knowledge Graph Embedding Framework for Fault Diagnosis in Railway Operational Equipment

Reliable fault diagnosis in railway operational equipment is critical to ensuring system safety, operational efficiency, and predictive maintenance. Existing methods struggle to capture the intricate interdependencies among fault causes, failure modes, and corrective actions, limiting their ability to model fault propagation effectively. To address this, we propose TH-RotatE, a novel knowledge graph (KG) embedding framework that integrates TransH’s hierarchical modeling with RotatE’s complex space transformations, while incorporating a hybrid scoring function and self-adversarial negative sampling to enhance embedding quality and fault relationship differentiation. This approach effectively captures hierarchical dependencies, cyclic patterns, and asymmetric transitions inherent in railway faults, enabling a more expressive representation of fault propagation. Furthermore, we construct the Chinese railway operational equipment fault knowledge graph (CROEFKG), a structured multi-relational repository encoding fault descriptions, causal chains, and mitigation strategies. Extensive experiments on real-world railway fault data demonstrate that TH-RotatE outperforms both traditional and advanced KG embedding models, achieving superior fault diagnosis accuracy and link prediction effectiveness. In practical applications, TH-RotatE enables timely fault diagnosis and detection of cascading failures, providing interpretable fault propagation pathways through the CROEFKG’s structured representation. These capabilities offer a scalable, knowledge-driven solution for railway systems, improving diagnostic accuracy while reducing safety risks and unplanned downtime. This work advances domain-specific KG embeddings, bridging the gap between theoretical innovation and industrial reliability.

A hybrid approach combining deep learning and signal processing for bearing fault diagnosis under imbalanced samples and multiple operating conditions

To enhance bearing fault diagnosis performance under various operating conditions, this paper proposes a hybrid approach based on generative adversarial networks (GANs), transfer learning, wavelet transform time-frequency representations, asymmetric convolutional networks, and the multi-head attention mechanism (MAC-MHA). Firstly, GANs are utilized to generate new bearing fault data to meet the model’s training requirements. Then, wavelet transform is applied to convert the bearing vibration signals into time-frequency representations, capturing the temporal evolution of frequency components. Next, an improved asymmetric convolutional network (MAC-MHA), combined with the multi-head attention mechanism, is employed to enhance the focus on key time-frequency features, further improving fault diagnosis accuracy. Considering the differences in operating conditions, transfer learning techniques are applied to facilitate knowledge transfer from the source domain to the target domain, thereby enhancing the model’s generalization ability. Experimental results demonstrate the effectiveness and robustness of the proposed method under various operating conditions. Finally, the proposed hybrid fault diagnosis approach is validated using the PADERBORN and CWRU datasets.

Named Entity Recognition in Mechanical Fault Diagnosis: A Template-Free Prompt Learning Approach

Intelligent prediction, accurate diagnosis, and efficient repair of mechanical equipment faults are critical for ensuring production safety and enhancing efficiency in industrial processes. However, data scarcity and recognition efficiency remain significant challenges in named entity recognition (NER) for mechanical equipment faults. To address these issues, this study proposes a novel NER method based on template-free prompt learning. The model is initialized with limited labeled data and then leverages hidden entity cues in unlabeled data to generate insightful hints, guiding the model through deep retraining. Experimental results demonstrate significant improvements in F1 scores: 29.54%, 22.34%, and 19.67% on the MEFD, CONLL03, and MIT-Movie datasets, respectively, as labeled samples increased from 5 shots to 50 shots. Furthermore, compared to template-based prompt learning methods, the proposed approach achieved F1 score improvements of 9.89%, 12.97%, 9.51%, and 2.21% as labeled samples scaled from 5 shots to 50 shots. The proposed method effectively mitigates data dependency issues, enhances model generalization and application capabilities, and provides robust technical support for the intelligent identification and diagnosis of mechanical equipment faults.

Bearing Fault Diagnosis Based on Multiscale Lightweight Convolutional Neural Network

Many bearing fault diagnosis methods often struggle to balance between adequate feature extraction and lightweight property, which makes it somewhat difficult to fulfill the accuracy and efficiency required for practical applications. To address this issue, this study describes the development of a multiscale lightweight deep learning model for accurate bearing fault diagnosis. Specifically, the Gaussian pyramid method, which can create a series of images at different scales, is employed to express the Gramian angular field (GAF) matrix images generated by transforming the bearing vibration signals to avoid the common problem of insufficient feature extraction of a single-scale image. At the same time, the dependencies between feature channels are extracted using a lightweight attention mechanism utilized in deep learning, known as Efficient Channel Attention (ECA), to improve the capability of feature representation. This approach effectively improves the learning ability of bearing fault characteristics and greatly increases the accuracy of fault diagnosis. Considering the problem related to the lightweight level of the method, a Ghost module, a type of convolution neural network system, is also employed to generate more features by using fewer parameters, thereby improving the overall calculation efficiency. Here we have developed a residual module based on the Ghost module and ECA, which can be easily integrated into most bearing fault diagnosis backbone networks. Based on our experimental tests, the developed system can clearly achieve high accuracy precision of bearing fault diagnosis to fulfill the needs of practical engineering while maintaining light weight. Specifically, the test accuracy of the proposed method using two bearing fault datasets exceeds 99.4%, and the giga floating-point operations (GFLOPs) is only 1.99, which can fully meet the needs of practical engineering.

Fault Diagnosis of Permanent Magnet Synchronous Motor Based on Wavelet Packet Transform and Genetic Algorithm-Optimized Back Propagation Neural Network

In this paper, a fault diagnosis method for permanent magnet synchronous motors is proposed, combining wavelet packet transform (WPT) energy feature extraction and a genetic algorithm (GA)-optimized back propagation (BP) neural network. Firstly, for the common types of motor faults (turn-to-turn short-circuit, phase-to-phase short-circuit, loss of magnetism, inverter open-circuit, rotor eccentricity), a corresponding motor fault model is established. The stator current signals during motor operation are analyzed using wavelet packet transform, and energy features are extracted from them as feature vectors for fault diagnosis. Then, a BP neural network is constructed, and a genetic algorithm is used to optimize its initial weights and thresholds, thereby improving the network’s classification accuracy. The results show that the GA-BP model outperforms the SSA-PNN diagnostic model in terms of fault classification accuracy. In particular, for the diagnosis of normal operation, inverter open-circuit, and demagnetization faults, the accuracy rate reaches 100%. This method demonstrates high diagnostic accuracy and practical application value.

An integrating multidimensional features method based on multi-view learning for intelligent fault diagnosis of rolling bearings

An integrating multidimensional features method based on multi-view learning for intelligent fault diagnosis of rolling bearings

Research on Rolling Bearing Fault Diagnosis Method Based on MPE and Multi-Strategy Improved Sparrow Search Algorithm Under Local Mean Decomposition

To address the issues of non-stationarity, noise interference, and insufficient discriminative power of traditional fault feature extraction methods in rolling bearing vibration signals, this paper proposes a fault diagnosis method based on multi-scale permutation entropy (MPE) and a multi-strategy improved sparrow search algorithm (MSSA) under local mean decomposition (LMD). First, LMD is employed to adaptively decompose the original signal. Effective product functions (PFs) are then selected using the Pearson correlation coefficient, enabling signal reconstruction that suppresses noise interference while preserving fault impact components. Second, to overcome the limited capability of traditional time-frequency features in representing complex fault patterns, MPE is introduced to construct a multi-scale complexity feature vector, effectively capturing the scale-dependent differences in the dynamic behavior of signals. Furthermore, considering the instability of classification caused by the empirical setting of hidden layer nodes in the extreme learning machine (ELM), a multi-strategy improved sparrow search algorithm is proposed to optimize ELM parameters. This algorithm integrates an adaptive Levy flight mechanism and dynamic reverse learning. The long-tail jump characteristics of Levy flight enhance the global search capability, while dynamic reverse learning increases population diversity, preventing premature convergence. The experimental results demonstrate that the proposed method achieves an average diagnostic accuracy of over 96% across multiple datasets, verifying its robustness in handling non-stationary signals and fault classification.

Clustering-guided prototypical contrastive learning for bearing fault diagnosis under variable working conditions

Abstract Methods based on deep learning for intelligent fault diagnosis have shown good results in general diagnostic tasks. Nevertheless, these methods largely depend on the sufficient labeled data, limiting their application in the actual scenarios where the availability of labeled data is limited. Moreover, the distribution of testing data is inconsistent with that of training data because bearings operate in various working conditions, leading to the performance degradation of these approaches. To tackle these two entangled problems, we propose a novel unsupervised domain adaptation network, which presents clustering-guided prototypical contrastive learning for cross-domain fault diagnosis. More specifically, k-means clustering is first used to aggregate similar source samples and target samples separately, acquiring the centroid of each cluster and the cluster index of each sample. Then, we propose in-domain and cross-domain contrastive learning strategies based on clustering results to achieve class alignment and domain alignment across source domain and target domain. By applying in-domain contrastive learning, we make the intra-class distance smaller while making the inter-class distance larger within each domain, effectively reducing the number of samples on the class boundaries. By applying cross-domain contrastive learning, class-to-class semantic similarity across two different domains is considered, which not only retains class discriminability in each domain but aligns these two domains at both the class level and the domain level. Detailed experiments on three bearing datasets reveal that our method outperforms in fault diagnosis across diverse working conditions, achieving average accuracy improvements of 2.10%, 7.44%, and 1.17% on the JNU, HUST, and Ottawa datasets, respectively.

Bearing Fault Diagnosis Method Based on Improved VMD and Parallel Hybrid Neural Network

In order to combat the difficulty of fault feature extraction and fault recognition in the field of bearing fault diagnosis, a bearing fault diagnosis method based on improved variational mode decomposition (VMD) and parallel hybrid neural network is proposed, which combines reweighted kurtosis (RK) with variable mode decomposition (VMD) and uses reweighted kurtosis as the evaluation index to select the decomposition times of variational mode decomposition, while removing part of the interference in the fault signal and retaining its impact characteristics. Afterwards, the processed fault data set is brought into a parallel hybrid neural network model with a global average pooling layer (GAP) for feature extraction, feature fusion, and fault classification. The parallel hybrid neural network model can extract fault signal features more comprehensively and improve the accuracy of fault diagnosis, while the global average pooling layer can speed up the training and testing. Experiments on the Xian Jiao tong University (XJTU) and Case Western Reserve University (CWRU) bearing public data sets show that the diagnosis accuracy reaches 99.72% and 99.73%, respectively, indicating that the method has good fault diagnosis accuracy and better diagnosis performance compared with other models.

Partially domain-adaptive rolling bearing fault diagnosis based on joint weighting and metric optimization

Abstract Partial-domain adaptive techniques are widely applied in cross-operational bearing fault diagnosis to address inconsistencies between source and target domain fault classes effectively. However, existing studies face challenges in feature alignment, including insufficient alignment of shared class fault features between the source and target domains, interference from outlier class samples in the source domain, and low-confidence pseudo-labels in the target domain. These issues hinder efficient alignment of shared class features, ultimately reducing fault diagnosis performance. This paper proposes a partial-domain deep migration model (JWMDA) that integrates joint weighting and metric optimization. To improve the alignment of shared class fault features, a joint metric combining covariance alignment (CORAL) and local maximum mean difference (LMMD) is developed. This metric complements adversarial training, reduces distributional differences between domains, and optimizes feature alignment. To address interference from outlier class samples in the source domain, class-level weights are employed to effectively mitigate the negative migration effects of these samples. Additionally, sample-level weights are introduced to reduce the negative migration effects of low-confidence pseudo-labels and boundary samples, enhancing the accuracy and robustness of shared class feature alignment. The proposed method is validated through experiments on both public and self-constructed bearing datasets. Experimental results demonstrate that the proposed method achieves higher diagnostic accuracy than existing partially domain-adaptive methods in cross-operational diagnostic tasks involving similar and different equipment.

Practical periodic feature enhancement framework for the detection of acoustic-based bearing faults

Abstract This paper proposes a two-step fault diagnosis scheme to address the challenge of weak fault feature extraction in bearing fault diagnosis using acoustic data. A practical periodic feature enhancement framework (PPFEF) is presented for deconvolving fault features. The method employs signal unbiased autocorrelation to optimize the deconvolution unified filter framework developed by synthesizing existing methods. Fault shock enhancement is achieved without needing previous fault information. Furthermore, to improve the effectiveness of PPFEF in extracting fault shocks, this paper presents an enhanced symplectic geometry modal decomposition (ESGMD) method for prenoise reduction. Inspired by singular value decomposition theory and dimension reduction concepts, the covariance matrix and trajectory matrix with sufficient essential features are built to strengthen the demodulation capability. A covariance-based discrete entropy is produced to characterize the fault periodic features and identify fault-sensitive components in the ESGMD. The effectiveness and key properties of PPFEF and ESGMD are analyzed. Numerical simulations and experimental studies are conducted on ESGMD–PPFEF.

Fault Diagnosis for Imbalanced Datasets Based on Deep Convolution Fuzzy System

Fault Diagnosis for Imbalanced Datasets Based on Deep Convolution Fuzzy System

A Sensor Data-Driven Fault Diagnosis Method for Automotive Transmission Gearboxes Based on Improved EEMD and CNN-BiLSTM

With the rapid development of new energy vehicle technologies, higher demands have been placed on fault diagnosis for automotive transmission gearboxes. To address the poor adaptability of traditional methods under complex operating conditions, this paper proposes a sensor data-driven fault diagnosis method based on improved ensemble empirical mode decomposition (EEMD) combined with convolutional neural networks (CNNs) and Bidirectional Long Short-Term Memory (BiLSTM) networks. The method incorporates a dynamic noise adjustment mechanism, allowing the noise amplitude to adapt to the characteristics of the signal. This improves the stability and accuracy of signal decomposition, effectively reducing the instability and error accumulation associated with fixed-amplitude white noise in traditional EEMD. By combining the CNN and BiLSTM modules, the approach achieves efficient feature extraction and dynamic modeling. First, vibration signals of the transmission gearbox under different operating states are collected via sensors, and an improved EEMD method is employed to decompose the signals, removing background noise and nonstationary components to extract diagnostically significant intrinsic mode functions (IMFs). Then, the CNN is utilized to extract features from the IMFs, deeply mining their spatiotemporal characteristics, while the BiLSTM captures the temporal sequence dependencies of the signals, enhancing the comprehensive modeling of nonlinear and dynamic fault features. The combination of these two networks enables efficient adaptation to complex conditions, achieving accurate classification and identification of multiple gearbox fault modes. Results indicate that the proposed approach is highly accurate and robust for identifying gearbox fault modes, significantly exceeding the performance of conventional methods and isolated network models. This provides an efficient and intelligent solution for fault diagnosis of automotive transmission gearboxes.

Advanced thermal vision techniques for enhanced fault diagnosis in electrical equipment: a review

Advanced thermal vision techniques for enhanced fault diagnosis in electrical equipment: a review

Sustainable Emission Control in Heavy-Duty Diesel Trucks: Fuzzy-Logic-Based Multi-Source Diagnostic Approach

Motor vehicles emit a large amount of air pollutants. Inspection and Maintenance (I/M) systems serve as a pivotal strategy for mitigating emissions from operational diesel trucks. However, the prevalent issue of blind repairs persists due to insufficient diagnostic capabilities at maintenance stations (M stations). To address this challenge, a multi-source information fusion methodology is proposed, integrating load deceleration testing from inspection stations (I stations), on-board diagnostics (OBD) data, and manual measurements at M stations. Critical diagnostic parameters—including nitrogen oxides (NOx) and particulate matter (PM) emissions, the ratio of measured wheel-side power to rated power, intake volume, common rail pressure, and exhaust back pressure—are systematically selected through statistical analysis and expert evaluations. An adaptive membership function is developed to resolve ambiguities in emission thresholds, enabling the construction of a robust fault diagnosis framework. Validation using 800 National V diesel truck maintenance records from a provincial automotive electronic health platform (2022 data) demonstrates a diagnostic accuracy of 92.8% for 153 emission-exceeding vehicles, surpassing traditional machine learning approaches by over 20%. By minimizing unnecessary repairs and optimizing maintenance efficiency, this approach significantly reduces resource waste and the lifecycle environmental footprints of diesel fleets. The proposed fuzzy-logic-based model effectively detects latent faults during routine maintenance, directly contributing to sustainable transportation through reductions in NOx and PM emissions—critical for improving air quality and advancing global climate objectives. This establishes a scalable technical framework for the effective implementation of I/M systems in alignment with sustainable urban mobility policies.