Accurate Fault Research Articles

Enhanced Fault Prediction for Synchronous Condensers Using LLM-Optimized Wavelet Packet Transformation

This paper presents an enhanced fault prediction framework for synchronous condensers in UHVDC transmission systems, integrating Large Language Models (LLMs) with optimized Wavelet Packet Transform (WPT) for improved diagnostic accuracy. The framework innovatively employs LLMs to automatically optimize WPT parameters, addressing the limitations of traditional manual parameter selection methods. By incorporating a Multi-Head Attention Gated Recurrent Unit (MHA-GRU) network, the system achieves superior temporal feature learning and fault pattern recognition. Through intelligent parameter optimization and advanced feature extraction, the LLM component intelligently selects optimal wavelet decomposition levels and frequency bands, while the MHA-GRU network processes the extracted features for accurate fault classification. Experimental results on a high-capacity synchronous condenser demonstrate the framework’s effectiveness in detecting rotor, air-gap, and stator faults across diverse operational conditions. The system maintains efficient real-time processing capabilities while significantly reducing false alarm rates compared to conventional methods. This comprehensive approach to fault prediction and diagnosis represents a significant advancement in synchronous condenser fault prediction, offering improved accuracy, reduced processing time, and enhanced reliability for UHVDC transmission system maintenance.

Interpretable Intelligent Fault Diagnosis for Heat Exchangers Based on SHAP and XGBoost

Heat exchangers play an important role in offshore oil and gas production and offshore low-temperature waste heat power generation. Heat exchanger health management mainly relies on performance parameter monitoring; however, accurate fault mode identification remains a challenge and the diagnostic results are difficult to interpret. This paper proposed an interpretable intelligent diagnosis method for heat exchangers based on Shapley Additive exPlanation and eXtreme Gradient Boosting. Firstly, secondary parameters characterizing the health state of heat exchangers are constructed based on online monitored data such as pressure and temperature. Then, the multi-dimensional feature vector is constructed by integrating the online monitored data and secondary parameters. The XGBoost fault diagnosis model for leakage and scaling faults is established with the multi-dimensional feature vector as the input. To improve the model performance on diagnosis, a grid search optimization algorithm is adopted. After the fault diagnosis of a heat exchanger is completed, the SHAP method is introduced to analyze the contribution of features by quantifying their influence on the fault diagnosis results. Finally, the effectiveness of the proposed method is verified by the heat exchanger fault simulation experiment. The proposed method achieved a diagnosis accuracy of 99.79%, which has better performance than the traditional fault diagnosis method.

DAE-BiLSTM Model for Accurate Diagnosis of Bearing Faults in Escalator Principal Drive Systems

The extensive deployment of escalators has greatly improved travel convenience; however, significant concerns have been raised due to the increasing frequency of safety incidents in recent years. Ensuring the safe operation of escalators and detecting faults in a timely manner have become critical concerns for both manufacturers and maintenance personnel. Traditional periodic inspections are resource-intensive and increasingly deemed inadequate due to the growing diversity and number of escalators. In this article, a data acquisition and transmission system for the main drive shaft bearing of the escalator, based on the Internet of Things (IoT), is designed using the main drive shaft bearing as an example. Additionally, a fault classification model combining a deep autoencoder (DAE) and Bidirectional Long Short-Term Memory Network (BiLSTM) is proposed. The experimental results of this study demonstrate that the DAE-BiLSTM-based fault diagnosis model provides accurate fault detection and early warnings, achieving an accuracy rate exceeding 99%, while significantly reducing the computational costs and training time.

Power Converter Fault Detection Using MLCA–SpikingShuffleNet

With the widespread adoption of electric vehicles, the power converter, as a key component, plays a crucial role. Traditional fault detection methods often face challenges in real-time performance and computational efficiency, making it difficult to meet the demands of electric vehicle power converters for efficient and accurate fault diagnosis. To address this challenge, this paper proposes a novel fault detection model—SpikingShuffleNet. This paper first designs an efficient SpikingShuffle Unit that integrates grouped convolutions and channel shuffle techniques, effectively reducing the model’s computational complexity by optimizing feature extraction and channel interaction. Next, by appropriately stacking SpikingShuffle Units and refining the network architecture, a complete lightweight diagnostic network is constructed for real-time fault detection in electric vehicle power converters. Finally, the Mixed Local Channel Attention mechanism is introduced to address the potential limitations in feature representation caused by grouped convolutions, further enhancing fault detection accuracy and robustness by balancing local detail preservation and global feature integration. Experimental results show that SpikingShuffleNet exhibits excellent accuracy and robustness in the fault detection task for power converters, fulfilling the real-time fault diagnosis requirements for low-power embedded devices.

Data sampling strategies for accurate fault analyses: A scale-independent test based on a Machine Learning approach

Data sampling strategies for accurate fault analyses: A scale-independent test based on a Machine Learning approach

Improving subsurface structural interpretation in complex geological settings through geophysical imaging and machine learning

This study employs seismic refraction tomography (SRT) and electrical resistivity tomography (ERT) to assess subsurface geological conditions along the proposed Porsgrunn Highway in Norway. The primary objective is to analyze SRT and ERT tomograms to identify subsurface geological structures. However, interpreting tomograms is often limited by smoothed boundaries and reduced resolution. To address these challenges, we apply k-means clustering, a machine learning technique that groups data based on similarities in physical properties, to post-process the geophysical tomograms. This study pioneers the use of k-means clustering to interpret tomograms from SRT and ERT data in complex geological settings. We first evaluate the effectiveness of clustering techniques using numerical modeling for two geological scenarios: a horizontally layered case and a layered case with undulation and a fault structure. Utilizing automated methods (Elbow and Silhouette), we objectively determine the optimal number of clusters for each geophysical tomogram. Subsequently, we compare the performance of the k-means clustering algorithm with subjective expert interpretations and the Laplacian edge detection method. Borehole data validate the clustering results and confirm the effectiveness of optimal cluster selection techniques. The findings of this study demonstrate that k-means clustering significantly enhances the detection of geological structures by establishing clearer boundaries and minimizing noise interference, enabling more accurate fault zone delineation. Compared to traditional edge detection and subjective interpretation methods, k-means clustering offers a systematic and objective approach that improves consistency and reliability across diverse geological settings. Moreover, its automated classification of geophysical data into meaningful clusters enables efficient analysis of large datasets. This study underscores the value of integrating machine learning techniques with geophysical methods such as SRT and ERT to improve interpretability and accurately identify subsurface geological structures, particularly in fault zone identification.

Few-shot fault diagnosis for machinery using multi-scale perception multi-level feature fusion image quadrant entropy

Aero-engines, pumps, and trains are widely used in transportation, maritime, aerospace, and other industries. However, these devices often operate in harsh and complex environments, making their internal components prone to failure. Thus, constructing a highly accurate fault diagnosis model is essential for ensuring the safe and reliable operation of machinery. However, most existing models require many labeled samples to build accurate training models, which is both expensive and difficult to achieve. Moreover, some models lack adaptability and often require adjustments to their structure or hyperparameters to suit new diagnostic tasks. To address these challenges, this paper proposes a few-shot fault diagnosis model based on multi-scale perception multi-level feature fusion image quadrant entropy (MPMFFIQE). The MPMFFIQE method uses the gramian angle summation field (GASF) to convert transient signals into images, preserving more detailed information about the mechanical state. The multi-scale perception multi-level feature strategy is then applied to sequentially enlarge and reconstruct feature maps at various levels, maximizing the extraction of fault-related information. Afterward, the fusion image quadrant entropy technique is proposed to combine nonlinear dynamic features from these feature maps, forming the mechanical MPMFFIQE feature set. Finally, this set is input into the harris hawks optimization support vector machine (HHOSVM) classifier to achieve fault identification. Results from three real-world case studies demonstrate that the proposed MPMFFIQE method improves accuracy by up to 12.90% in comparison with six feature extraction techniques. Furthermore, the proposed model achieves an accuracy rate exceeding 98.10% with just five training samples per state, representing up to a 27.48% improvement over six existing models. These findings confirm that the developed model can effectively and accurately diagnose mechanical faults in industrial applications using only a small number of training samples. Additionally, the model shows strong generalization across different mechanical equipment, highlighting its significant practical value.

Unified Flowing Normality Learning for Rotating Machinery Anomaly Detection in Continuous Time-Varying Conditions.

Intelligent anomaly detection (AD) methods have achieved much successes in machinery condition monitoring. However, the underlying independent and identically distributed assumption restricts their application scopes to steady operating conditions. False and missing alarms would occur when machines operate under time-varying circumstances. In this work, a more challenging time-varying setting is studied, where the working conditions are continuously changing, such that few or no samples are available for model training at one single condition. To tackle this issue, we propose a unified flowing normality learning (UFNL) framework, which aims to capture the flowing normal conditional distribution of time-varying samples and assigns dynamic decision boundary for AD. Specifically, a manifold-based probability density estimation is utilized to guide the adversarial learning process of generative adversarial networks, where adjacent samples are aggregated to approximate the conditional distribution by a conditional generator. Then, a latent normality inversion is proposed to extract the manifold structure from the pretrained generator and to map it into the latent space via a conditional encoder. The reconstruction errors from the encoder and generator can reveal the deviation of signals to the flowing normality. Finally, a condition-aware adaptive threshold selection strategy is proposed, where different thresholds are adaptively assigned for different conditions. Experiments are carried out under two typical continuous time-varying scenarios. The results demonstrate that the proposed framework can realize accurate fault detection at any operating condition within continuously changing environments.

A Deployment Method for Motor Fault Diagnosis Application Based on Edge Intelligence

The rapid advancement of Industry 4.0 and intelligent manufacturing has elevated the demands for fault diagnosis in servo motors. Traditional diagnostic methods, which rely heavily on handcrafted features and expert knowledge, struggle to achieve efficient fault identification in complex industrial environments, particularly when faced with real-time performance and accuracy limitations. This paper proposes a novel fault diagnosis approach integrating multi-scale convolutional neural networks (MSCNNs), long short-term memory networks (LSTM), and attention mechanisms to address these challenges. Furthermore, the proposed method is optimized for deployment on resource-constrained edge devices through knowledge distillation and model quantization. This approach significantly reduces the computational complexity of the model while maintaining high diagnostic accuracy, making it well suited for edge nodes in industrial IoT scenarios. Experimental results demonstrate that the method achieves efficient and accurate servo motor fault diagnosis on edge devices with excellent accuracy and inference speed.

Classification and fault diagnosis of power transformers with dissolved gas analysis using improved clustering methods

Purpose Power transformers are vital components of an electrical network. A defective transformer can cause instability and blackouts in parts of the network. An accurate classification of different transformer faults results in a relatively accurate fault diagnosis and timely corrective actions. It is possible to increase productivity and reduce costs by using fault detection of power transformers through the analysis of gases dissolved in oil. The proposed technique is a suitable tool to help the utilities and engineers in charge of preventive maintenance by reducing the costs of different fault diagnosis tests for power transformers. Design/methodology/approach In this paper, the IEC 60599 standard along with clustering and classification methods are used to classify power transformer’s fault types. K-means and Fuzzy C-means clustering methods are used for clustering, and the support vector machine (SVM) method is used for classification of different types of faults in ‎power ‎transformers. The performance of K-means and SVM methods is improved by using the Grasshopper Optimization Algorithm (GOA). The efficiency of the proposed methods is evaluated using real field data of power transformers. The purpose of this study is to propose hybrid methods including K-means-GOA clustering and SVM-GOA classification for accurate fault diagnosis. These methods have been used for the first time in fault diagnosis determination of power transformers through gas analysis. The Silhouette criteria is used in this paper to compare the efficiency of different clustering methods. Findings Simulation results of the paper are based on the gas chromatography data related to 266 different real power transformers. They show the high accuracy and high-performance speed of intelligent clustering and classification methods compared to conventional ones. This analysis would be helpful in performing the required maintenance check and plan for repairs. Originality/value The applicability and efficiency of the proposed hybrid K-means-GOA and SVM-GOA models are verified for transformer fault detection using the experimental diverse data set including 266 set of real field test parameters of power transformers.

A novel cross domain diagnosis method based on physical feature weighting and deep residual shrinkage network

Abstract Traditional intelligent diagnostic methods often exhibit poor generalization ability when faced with data from different experimental platforms due to their differing distributions. Current transfer learning methods typically require fine-tuning of the model with some target domain data to adapt to different data distributions. However, in practical applications, it is often difficult to obtain sufficient target domain data. To address this issue, this paper proposes an innovative cross-domain diagnostic method that can diagnose faults on a target platform without relying on its data. The method introduces fault information of bearings into the attention weights of a Deep Residual Shrinkage Network (DRSN) through a novel physical feature weighting layer. This effectively extracts the hidden fault information from the signals, which is further refined and classified by DRSN to obtain an accurate fault diagnosis result. Experiments conducted on three bearing datasets validate the effectiveness of the proposed method, demonstrating its promising application prospects for bearing fault diagnosis under different operating conditions.

Fault analysis and elimination of motor sensors in pure electric vehicles

Sensor as the main component of the motor drive system is to ensure its stable and reliable operation of the core components, so to drive the system's normal operation must learn how to analyse and diagnose the motor sensor fault accurately. This paper first introduces different fault classification and basic diagnosis methods, uses the CNN algorithm to collect sensor information, and uses matrix, dot product and other mathematical operations to make deep diagnosis. The sensor's real-time working status and accurate fault feedback can be obtained. CNN algorithm is fully suitable for the research and application of motor sensor fault diagnosis and can provide the basis for fault diagnosis according to the change of its own operating data parameters, and achieve efficient analysis and elimination of existing faults, to a certain extent, it can provide reference for the replacement of motor sensors and provide analysis for personnel engaged in related diagnosis industries.

Hybrid Deep Learning Model for Fault Diagnosis in Centrifugal Pumps: A Comparative Study of VGG16, ResNet50, and Wavelet Coherence Analysis

Significant in various industrial applications, centrifugal pumps (CPs) play an important role in ensuring operational efficiency, yet they are susceptible to faults that can disrupt production and increase maintenance costs. This study proposes a robust hybrid model for accurate fault detection and classification in CPs, integrating Wavelet Coherence Analysis (WCA) with deep learning architectures VGG16 and ResNet50. WCA is initially applied to vibration signals, creating time–frequency representations that capture both temporal and frequency information, essential for identifying subtle fault characteristics. These enhanced signals are processed by VGG16 and ResNet50, each contributing unique and complementary features that enhance feature representation. The hybrid approach fuses the extracted features, resulting in a more discriminative feature set that optimizes class separation. The proposed model achieved a test accuracy of 96.39%, demonstrating minimal class overlap in t-SNE plots and a precise confusion matrix. When compared to the ResNet50-based and VGG16-based models from previous studies, which reached 91.57% and 92.77% accuracy, respectively, the hybrid model displayed better classification performance, particularly in distinguishing closely related fault classes. High F1-scores across all fault categories further validate its effectiveness. This work underscores the value of combining multiple CNN architectures with advanced signal processing for reliable fault diagnosis, improving accuracy in real-world CP applications.

Early Fault Diagnosis and Prediction of Marine Large-Capacity Batteries Based on Real Data

The inconsistency of battery voltages in all-electric ships is a significant issue for electric vehicle battery systems, leading to numerous safety concerns during vessel operation. Therefore, timely fault diagnosis and accurate fault prediction are crucial for the safe operation of ships. This study examines the fault alarm system of marine battery management systems in conjunction with the unique operating conditions of ships, focusing on the system’s latency. To facilitate prompt fault detection, a fault diagnosis method based on the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm is proposed, utilizing the voltage data of battery clusters. Results indicate that the DBSCAN clustering algorithm demonstrates superior effectiveness and accuracy in identifying irregular battery clusters. Furthermore, the fault prediction method based on the iTransformer model is introduced to forecast variations in battery cluster voltages. Experimental findings suggest that this model can effectively predict consistency faults and over-/under-voltage conditions based on battery cluster voltage values and corresponding fault thresholds.

Cable fault diagnosis with generalization capability using incremental learning and deep convolutional neural network

Power cable faults may exhibit dynamic and differential changes attributable to variations in operating times and environmental conditions. However, the existing artificial intelligence (AI)-based fault diagnosis methods are lack of capability to handle this complexity of fault development. They can only diagnose given faulty types and will malfunction when dealing with newly-emerged faults, which accounts for the lack of generalization capability for AI-based methods. Given that, this paper proposes a cable fault diagnosis method with generalization capability utilizing incremental learning and deep convolutional neural network (DCNN). The idea of knowledge distillation is introduced to reserve the diagnosis information of the DCNN for original faults, and meanwhile, the DCNN is also modified by the classification loss of newly-emerged faulty samples, which equips the DCNN with generalization capability. The category features are obtained utilizing the refined DCNN, and sent to the constructed classifier for accurate incremental fault diagnosis. Experiments are conducted with field and simulated data. The obtained results validate the feasibility and effectiveness of the proposed method.

Quantitative Evaluation for Robustness of Intelligent Fault Diagnosis Algorithms based on Self-attention Mechanism

Currently, various algorithmic models encounter numerous challenges in practical applications, such as noise, interference, and input changes, which can significantly impact their performance. Many methods have been proposed to enhance model robustness. However, to assess the effectiveness of these improvements, it is generally necessary to compare the model’s performance before and after applying the same noise and analyze the resulting changes. Moreover, evaluating the robustness of multiple models that meet basic requirements for a specific task requires qualitative analysis using specific indicators. This is especially crucial in fault diagnosis where multiple types of noise interference in the data can hinder accurate fault classification. Addressing this situation, this paper presents a quantitative evaluation method for the robustness of intelligent fault diagnosis algorithms based on the self-attention mechanism. The proposed method entails dividing the dataset into sub-datasets according to signal-to-noise ratio after injecting noise, separately calculating sub-indicators after training, dynamically assigning weights to these indicators using the self-attention mechanism and combining the weights of different sub-indicators to generate a comprehensive evaluation value for assessing robustness. The proposed method was validated through experiments involving three models, and the results demonstrated the reliability of this quantitative calculation approach for robustness.

A robust diagnostic approach in mean shifts for multivariate statistical process control

The control chart is very useful and efficient to detect abnormal changes of a process, however it is difficult to identify out-of-control variables that are responsible for the change. Accurate fault diagnosis has become increasingly important, which helps engineer quickly eliminate the root causes of abnormal changes. However, most of existing diagnostic methods were developed under the assumption of normality, of which the performance is affected by the nonnormality. To overcome these limitations, this paper attempts to combine Bootstrap technique to propose a diagnosis method that is robust to underlying distribution, but also applicable to the case of small out-of-control sample size. Bootstrap technique is used to improve the diagnostic ability of the diagnostic method, which is our first attempt. Compared with the existing procedures, numerical simulations favour the proposed procedure. Finally, an example from a bolt production process is provided to illustrate the implementation of the proposed procedure.

ResNest-SVM-based method for identifying single-phase ground faults in active distribution networks

Single-phase grounding fault is the most common fault type in the distribution network. An accurate and effective single-phase grounding fault identification method is a prerequisite for maintaining the safe and stable operation of the power grid. Most neutral points of the active distribution network are grounded through arc suppression coils. In the active distribution network, the power supply in the network changes from one to multiple, which may change the direction of the fault current. In this paper, the superposition theorem is used to analyze the difference in the boosting effect of different types of distributed generators (DG) on line mode current in the sequence network diagram when DG is connected upstream or downstream of the fault point. Secondly, the composition of the zero-mode transient current of the fault line is analyzed. A judgment method based on the superposition diagram of transient zero-sequence voltage and current is proposed. Then, this paper improves the ResNest network and modifies the classifier of the last fully connected layer to SVM. Finally, the model in PSCAD is used to simulate single-phase grounding faults to obtain the training set and validation set. These datasets are used to train and test AlexNet, ResNet50, ResNeSt, and ResNeSt-SVM. The results show that under different fault points, transition resistances, DG access upstream and downstream of the fault point, and different fault initial phase angles, the ResNest-SVM model method can accurately identify the fault line and has better anti-noise ability than the other three network structures.

Enhanced Fault Detection in Satellite Attitude Control Systems Using LSTM-Based Deep Learning and Redundant Reaction Wheels

Reliable fault detection in satellite attitude control systems stands as a critical aspect of ensuring the safety and success of space missions. Central to these systems, reaction wheels (RWs), despite being the most frequently used actuators, present a vulnerability given their susceptibility to faults—a factor with the potential to precipitate catastrophic failures such as total satellite loss. In light of this, we introduce a fault detection methodology grounded in deep learning techniques specifically designed for satellite attitude control systems. Our proposed method utilizes a Long Short-Term Memory (LSTM) model adept at learning temporal patterns inherent to both healthy and faulty system behaviors. Incorporated into our model is a torque allocation algorithm designed to circumvent specific velocities known to induce torque disturbances, a factor known to influence LSTM performance adversely. To bolster the robustness of our fault detection technique, we also incorporated denoising autoencoders within the LSTM framework, thereby enabling the model to identify temporal patterns in healthy and faulty system behavior, even amidst the noise. The method was evaluated using cross-validation on simulated satellite data comprising 1000 time series samples and across different fault scenarios, such as stiction and resonance at varying intensities (90%, 50%, and 30%). The results confirm achieving performance metrics such as Mean Squared Error for accurate fault identification. This research underscores a stride in the evolution of fault detection and control strategies for satellite attitude control systems, holding promise to boost the reliability and efficiency of future space missions.

Research on Fault Diagnosis of Agricultural IoT Sensors Based on Improved Dung Beetle Optimization–Support Vector Machine

The accuracy of data perception in Internet of Things (IoT) systems is fundamental to achieving scientific decision-making and intelligent control. Given the frequent occurrence of sensor failures in complex environments, a rapid and accurate fault diagnosis and handling mechanism is crucial for ensuring the stable operation of the system. Addressing the challenges of insufficient feature extraction and sparse sample data that lead to low fault diagnosis accuracy, this study explores the construction of a fault diagnosis model tailored for agricultural sensors, with the aim of accurately identifying and analyzing various sensor fault modes, including but not limited to bias, drift, accuracy degradation, and complete failure. This study proposes an improved dung beetle optimization–support vector machine (IDBO-SVM) diagnostic model, leveraging the optimization capabilities of the former to finely tune the parameters of the Support Vector Machine (SVM) to enhance fault recognition under conditions of limited sample data. Case analyses were conducted using temperature and humidity sensors in air and soil, with comprehensive performance comparisons made against mainstream algorithms such as the Backpropagation (BP) neural network, Sparrow Search Algorithm–Support Vector Machine (SSA-SVM), and Elman neural network. The results demonstrate that the proposed model achieved an average diagnostic accuracy of 94.91%, significantly outperforming other comparative models. This finding fully validates the model’s potential in enhancing the stability and reliability of control systems. The research results not only provide new ideas and methods for fault diagnosis in IoT systems but also lay a foundation for achieving more precise, efficient intelligent control and scientific decision-making.