Search Results (180)

Most of the failures during the actual operation of equipment are caused by improper human operation, tools, spare parts, and environmental factors. These faults are routine. Conventional faults have been validated during equipment development, testing, identification, and maintenance processes, with clear definitions and clear fault tree analysis (FTA) conclusions. Digital twins can offer rapid and interactive diagnostic capabilities for routine equipment failures. To enhance the efficiency of routine fault diagnosis and the interactive experience of the diagnosis process, this paper proposes a digital twin-based equipment routine fault diagnosis model. On this basis, considering the excellent interactivity of the Interactive Electronic Technical Manual (IETM), a conventional equipment fault diagnosis scheme based on twin data and IETM is designed. This scheme converts the equipment fault tree into an IETM fault data model (DM), which is structured and stored in a database to form a fault database. Using real-time twin data of equipment as input, the FTA method is adopted to perform step-by-step fault diagnosis and isolation guidance operation through the IETM process DM combined with fault, while providing maintenance operation guidance. When the real-time twin data of the equipment is not completely consistent with the fault information in the fault library, the case analysis method is used to calculate the similarity between the real-time twin data of the equipment and the clearly defined fault symptom information in the fault library. Based on the set similarity threshold, IETM pushes fault DMs above the threshold for corresponding fault diagnosis isolation guidance. Full article

Series arc faults are a major cause of electrical fires, posing significant risks to life and property. Their negative-resistance characteristics make fault features difficult to detect, and the existing methods often suffer from high false-alarm rates, poor adaptability, and reliance on high sampling rates and long sampling windows. To enhance the accuracy and efficiency of series AC arc fault detection, this paper proposes a rapid identification method based on singular spectrum statistical features and a differential evolution-optimized XGBoost classifier. The approach first constructs the singular spectrum of current waveforms via a Hankel matrix singular value decomposition and extracts nine statistical features. It then optimizes seven XGBoost hyperparameters using differential evolution to build an efficient classification model. The experiments on 18,240 current samples covering 16 load conditions (including eight arc fault types) show that the method achieves an average identification accuracy of 98.90% using only three nominal cycles (60 ms) of current waveform. Even with a training set ratio as low as 5%, it maintains 97.11% accuracy, outperforming Back-propagation Neural Network, Support Vector Machine, and Recurrent Neural Network methods by up to three percentage points. The method avoids the need for high sampling rates or complex time–frequency transformations, making it suitable for resource-constrained embedded platforms and offering a generalizable solution for series arc fault detection. Full article

To address the issues of reduced operational efficiency, shortened equipment lifespan, and significant safety hazards caused by bearing wear and blade cavitation in hydroelectric turbine generators due to prolonged high-load operation, this paper proposes a ResNet + self-attention-based acoustic fingerprint fault diagnosis algorithm for hydroelectric turbine generators. First, to address the issue of severe noise interference in acoustic signature signals, the ensemble empirical mode decomposition (EEMD) is employed to decompose the original signal into multiple intrinsic mode function (IMF) components. By calculating the correlation coefficients between each IMF component and the original signal, effective components are selected while noise components are removed to enhance the signal-to-noise ratio; Second, a fault identification network based on ResNet + self-attention fusion is constructed. The residual structure of ResNet is used to extract features from the acoustic signature signal, while the self-attention mechanism is introduced to focus the model on fault-sensitive regions, thereby enhancing feature representation capabilities. Finally, to address the challenge of model hyperparameter optimization, a Bayesian optimization algorithm is employed to accelerate model convergence and improve diagnostic performance. Experiments were conducted in the real working environment of a pumped-storage power station in Zhejiang Province, China. The results show that the algorithm significantly outperforms traditional methods in both single-fault and mixed-fault identification, achieving a fault identification accuracy rate of 99.4% on the test set. It maintains high accuracy even in real-world scenarios with superimposed noise and environmental sounds, fully validating its generalization capability and interference resistance, and providing effective technical support for the intelligent maintenance of hydroelectric generator units. Full article

In order to improve the inversion accuracy of electrical resistivity tomography (ERT) and overcome the limitations of traditional linear methods, this paper proposes an improved shuffled frog leaping algorithm (SFLA). First, an equilibrium grouping strategy is designed to balance the contribution weight of each subgroup to the global optimal solution, suppressing the local optimum traps caused by the dominance of high-quality groups. Second, an adaptive movement operator is constructed to dynamically regulate the step size of the search, enhancing the guiding effect of the optimal solution. In synthetic data tests of three typical electrical models, including a high-resistivity anomaly with 5% random noise, a normal fault, and a reverse fault, the improved algorithm shows an approximately 2.3 times higher accuracy in boundary identification of the anomaly body compared to the least squares (LS) method and standard SFLA. Additionally, the root mean square error is reduced by 57%. In the engineering validation at the Baota Mountain mining area in Jurong, the improved SFLA inversion clearly reveals the undulating bedrock morphology. At a measuring point 55 m along the profile, the bedrock depth is 14.05 m (ZK3 verification value 12.0 m, error 17%), and at 96 m, the depth is 6.9 m (ZK2 verification value 6.7 m, error 3.0%). The characteristic of deeper bedrock to the south and shallower to the north is highly consistent with the terrain and drilling data (RMSE = 1.053). This algorithm provides reliable technical support for precise detection of complex geological structures using ERT. Full article

The accurate identification of seismic faults, which serve as crucial fluid migration pathways in hydrocarbon reservoirs, is of paramount importance for reservoir characterization. Traditional interpretation is inefficient. It also struggles with complex geometries, failing to meet the current exploration demands. Deep learning boosts fault identification significantly but struggles with edge accuracy and noise robustness. To overcome these limitations, this research introduces SwiftSeis-AWNet, a novel lightweight and high-precision network. The network is based on an optimized MedNeXt architecture for better fault edge detection. To address the noise from simple feature fusion, a Semantics and Detail Infusion (SDI) module is integrated. Since the Hadamard product in SDI can cause information loss, we engineer an Attention-Weighted Semantics and Detail Infusion (AWSDI) module that uses dynamic multi-scale feature fusion to preserve details. Validation on field seismic datasets from the Netherlands F3 and New Zealand Kerry blocks shows that SwiftSeis-AWNet mitigates challenges like the loss of small-scale fault features and misidentification of fault intersection zones, enhancing the accuracy and geological reliability of automated fault identification. Full article

With the rapid advancement of digitalization and intelligence in power systems, SF₆ high-voltage circuit breakers, as the core switching devices in power grid protection systems, have become critical components in high-voltage networks of 110 kV and above due to their superior insulation performance and exceptional arc-quenching capability. Their operational status directly impacts the reliability of power system protection. Therefore, real-time condition monitoring and accurate assessment of SF₆ circuit breakers along with science-based maintenance strategies derived from evaluation results hold significant engineering value for ensuring secure and stable grid operation and preventing major failures. In recent years, the frequency of extreme weather events has been increasing, necessitating a comprehensive consideration of both internal and external factors in the operational status prediction of SF₆ high-voltage circuit breakers. To address this, we propose an operational status assessment model for SF₆ high-voltage circuit breakers based on an Integrated Attribute-Weighted Risk Model Based on the Branch–Trunk Rule (IAR-BTR), which integrates internal and environmental influences. Firstly, to tackle the issues of incomplete data and feature imbalance caused by irrelevant attributes, this study employs missing value elimination (Drop method) on the fault record database. The selected dataset is then normalized according to the input feature matrix. Secondly, conventional risk factors are extracted using traditional association rule mining techniques. To improve the accuracy of these rules, the filtering thresholds and association metrics are refined based on seasonal distribution and the importance of time periods. This allows for the identification of spatiotemporally non-stationary factors that are strongly correlated with circuit breaker failures in low-probability seasonal conditions. Finally, a quantitative weighting method is developed for analyzing branch-trunk rules to accurately assess the impact of various factors on the overall stability of the circuit breaker. The DFP-Growth algorithm is applied to enhance the computational efficiency of the model. The case study results demonstrate that the proposed method achieves exceptional accuracy (95.78%) and precision (97.22%) and significantly improves the predictive performance of SF₆ high-voltage circuit breaker operational condition assessments. Full article

Photovoltaic (PV) systems are prone to different types of faults, primarily electrical faults such as line-to-ground (L-G) and line-to-line (L-L) faults, which can significantly reduce system performance, efficiency, and lead to increased power losses. Moreover, mechanical damage caused by environmental stressors (such as wind, hail, or temperature variations), aging, or improper installation also contribute to system degradation. This study specifically focuses on electrical faults and proposes a method that not only enables the isolation of faulty modules but also ensures the uninterrupted operation of the remaining healthy modules and also assists in the localization of faults. Unlike benchmarked techniques-based boost converters, the Quasi-Z-Source Boost Converter (QZBC) topology offers improved voltage boosting with high gain values, reduced component stress, and enhanced reliability when the PV system is undergoing fault identification and localization algorithms. A 600-watt PV system connected with a Quasi-Z-Source Boost Converter was implemented and tested under different fault conditions using a hardware-in-the-loop (HIL) setup with Typhoon HIL. All the component values of the QZBC were calculated based on the system requirements rather than assumed, ensuring both practical feasibility and design accuracy. The experimental results show that the converter achieved an efficiency of over 96% under electrical-fault conditions, confirming the effectiveness of the quasi-Z-source boost converter in maintaining a stable power output when the PV system is undergoing fault identification and localization algorithms. The study further highlights the benefits of HIL-based testing for evaluating PV-system resilience and fault-handling capabilities in real-time conditions using a Typhoon HIL 404 environment. Full article

The analysis and accurate identification of DC-side grounding faults in grid-connected photovoltaic (PV) inverters is a critical step in enhancing operation and maintenance capabilities and ensuring the safe operation of PV grid-connected systems. However, the characteristics of DC-side grounding faults remain unclear, and effective methods for identifying such faults are lacking. To address the need for leakage characteristic analysis and fault identification of DC-side grounding faults in grid-connected PV inverters, this paper first establishes an equivalent analysis model for DC-side grounding faults in three-phase grid-connected inverters. The formation mechanism and frequency-domain characteristics of residual current under DC-side fault conditions are analyzed, and the specific causes of different frequency components in the residual current are identified. Based on the leakage current mechanisms and statistical characteristics of grid-connected PV inverters, a multi-type DC-side grounding fault identification method is proposed using the light gradient-boosting machine (LGBM) algorithm. In the simulation case study, the proposed fault identification method, which combines mechanism characteristics and statistical characteristics, achieved an accuracy rate of 99%, which was significantly superior to traditional methods based solely on statistical characteristics and other machine learning algorithms. Real-time simulation verification shows that introducing mechanism-based features into grid-connected photovoltaic inverters can significantly improve the accuracy of identifying grounding faults on the DC side. Full article

During transformer operation, long-term vibration causes the winding to loosen axially. When hit by a short-circuit, the winding deforms to different extents. Thus, identifying early looseness faults in transformer windings is vital for power systems’ stability. To address issues including scarce vibration data across multiple load conditions for transformer winding looseness faults, inadequate extraction of two-dimensional spectrogram features, and the inability to boost recognition accuracy caused by overfitting during fault recognition model training, this study constructed a 10 kV power transformer vibration test platform. It measured the vibration signals on the box surface under various winding looseness conditions and built a time–frequency-domain vibration spectrum library for different load currents. Then, a fault identification model based on vibration spectra and ConvNeXt was constructed, and model verification and analysis were carried out. The results indicate that after training, the fault recognition accuracy of the spectrum containing three load conditions is comparable to that of a single load condition. The average recognition accuracy at six box-surface measuring points reaches 97.9%. Moreover, the ConvNeXt model outperforms the traditional ResNet50 by 1.2%. This new model effectively addresses overfitting and offers strong technical support for detecting different transformer winding looseness faults. Full article

To address the limited generalization capability of data-driven fault diagnosis models caused by scarce gearbox fault samples in engineering practice, this paper proposes a hierarchical attention-guided data–knowledge dual-driven fusion network for intelligent fault diagnosis under few-shot conditions. Distinct from traditional single data-driven paradigms, this method breaks through the constraints of limited samples through the synergy of prior knowledge and monitoring data. First, domain knowledge of gearbox fault diagnosis is utilized to construct prior features of monitoring data. Second, a deep convolutional neural network is designed to hierarchically capture abstract features from monitoring data. Subsequently, a hierarchical attention module is proposed to realize adaptive fusion of prior features and abstract features through hierarchical feature weight allocation, generating highly discriminative fused features for accurate gearbox fault identification. Experimental results on gearbox fault data demonstrate that the proposed method achieves 0.9880 recognition accuracy with less than 10% of the training samples, significantly outperforming purely data-driven models such as MGAN and CNET, thus verifying its superior generalization ability to train despite data scarcity. This approach establishes a novel data–knowledge dual-driven fusion paradigm for intelligent fault diagnosis of mechanical equipment under few-shot conditions. Full article

To enable proactive prediction of marine diesel engine failure time and root causes, thereby reserving sufficient time for maintenance, this study proposes a data-driven multi-algorithm integration framework for performance assessment and fault early warning in marine diesel engines. By integrating the SSD (steady-state detection) algorithm, a data-driven CLIQUE clustering algorithm was chosen for automatic multi-parameter high-dimensional running condition partitioning. This innovative approach overcomes the limitations of traditional single-parameter approaches or dimensionality reduction techniques, significantly enhancing state classification accuracy. The improved classification results subsequently increase the reliability of Mahalanobis distance as a performance indicator for marine diesel engine condition assessment. Finally, the cumulative anomaly method combined with the Yamamoto test was employed for anomaly detection analysis, enabling precise identification of fault occurrence time and establishing an effective early-warning mechanism. The study demonstrates that this technique effectively characterizes the overall performance of marine diesel engines and captures their performance degradation features. Implemented on a 6RT-flex82T marine diesel engine dataset, the method achieved precise prediction of fault occurrence time with early warnings, providing approximately 20 days advance notice for maintenance planning. Furthermore, comparative analyses with existing studies revealed its superior capability in pinpointing the anomaly to the jacket cooling water outlet temperature of cylinder #2. These results confirm the method’s effectiveness in both performance assessment and fault early warning for marine diesel engines, offering a novel approach for intelligent maintenance of shipboard equipment. Full article

Strong earthquake activity along fault zones can lead to the displacement of geomorphic units such as gullies and terraces while preserving earthquake event data through changes in sedimentary records near faults. The quantitative analysis of these characteristics facilitates the reconstruction of significant earthquake activity history along the fault zone. Recent advancements in acquisition technology for high-precision and high-resolution topographic data have enabled more precise identification of displacements caused by fault activity, allowing for a quantitative assessment of the characteristics of strong earthquakes on faults. The 1920 Haiyuan earthquake, which occurred on the Haiyuan fault in the northeastern Tibetan Plateau, resulted in a surface rupture zone extending nearly 240 km. Although clear traces of surface rupture have been well preserved along the fault, debate regarding the maximum displacement is ongoing. In this study, we focused on two typical offset geomorphic sites along the middle segment of the Haiyuan fault that were previously identified as having experienced the maximum displacement during the Haiyuan earthquake. High-precision geomorphologic images of the two sites were obtained through unmanned aerial vehicle (UAV) surveys, which were combined with light detection and ranging (LiDAR) data along the fault zone. Our findings revealed that the maximum horizontal displacement of the Haiyuan earthquake at the Shikaguan site was approximately 5 m, whereas, at the Tangjiapo site, it was approximately 6 m. A cumulative offset probability distribution (COPD) analysis of high-density fault displacement measurements along the ruptures indicated that the smallest offset clusters on either side of the Ganyanchi Basin were 4.5 and 5.1 m long. This analysis further indicated that the average horizontal displacements of the Haiyuan earthquake were approximately 4–6 m. Further examination of multiple gullies and geomorphic unit displacements at the Shikatougou site, along with a detailed COPD analysis of dense displacement measurements within a specified range on both sides, demonstrated that the cumulative displacement within 30 m of this section of the Haiyuan fault exhibited at least five distinct displacement clusters. These dates may represent the results of five strong earthquake events in this fault segment over the past 10,000–13,000 years. The estimated magnitude, derived from the relationship between displacement and magnitude, ranged from M_w 7.4 to 7.6, with an uneven recurrence interval of approximately 2500–3200 years. Full article

Inter-turn short circuits (ITSCs) in permanent magnet synchronous motors (PMSMs) pose significant risks due to their subtle early symptoms and rapid degradation. To address this, we propose an online real-time diagnostic method for assessing the degradation state. This method employs the Sparrow Search Algorithm (SSA) for the online real-time identification of fault characteristic parameters. Following an analysis of the fault mechanisms of inter-turn short circuits, a mathematical model has been developed to include the short-circuit turns ratio and insulation resistance. An evaluation index has also been developed to assess the degree of fault-related degradation. To address the strong nonlinearity of parameters in the fault model, the SSA is employed for the real-time joint identification of parameters that characterize the relationship between fault location and degradation degree. Simulation experiments demonstrate that the SSA achieves convergence within 40 iterations, with a relative error below 5% and absolute error less than 0.007, outperforming traditional algorithms like the PSO, a significant improvement in the early detection of degradation caused by inter-turn short circuits and a step forward in technical support ensuring greater reliability and safety for the traction systems used in rail transit. Full article

The contact condition of circuit breaker contacts directly affects their operational reliability, while circuit resistance, as a key performance indicator, reflects physical changes such as wear, oxidation, and ablation. Traditional offline measurement methods fail to accurately represent the real-time operating state of equipment due to large errors and high randomness, limiting their effectiveness for state awareness and precision maintenance. To address this, a non-contact multi-sensor fusion method for the online monitoring of circuit breaker circuit resistance is proposed, aimed at enhancing operational state awareness in power systems. The method integrates Hall effect current sensors, infrared temperature sensors, and electric field sensors to extract multiple sensing signals, combined with high-precision signal processing algorithms to enable the real-time identification and evaluation of circuit resistance changes. Experimental validation under various scenarios—including normal load, overload impact, and high-temperature and high-humidity environments—demonstrates excellent system performance, with a fast response time (≤200 ms), low measurement error (<1.5%), and strong anti-interference capability (SNR > 60 dB). In field applications, the system successfully identifies circuit resistance increases caused by contact oxidation and issues early warnings, thereby preventing unplanned outages and demonstrating a strong potential for application in the fault prediction and intelligent maintenance of power grids. Full article

Transformer winding faults (TWFs) can lead to insulation breakdown, internal short circuits, and catastrophic transformer failure. Due to their low current magnitude—particularly at early stages such as inter-turn short circuits, axial or radial displacement, or winding looseness—TWFs often induce minimal impedance changes and generate fault currents that remain within normal operating thresholds. As a result, conventional protection schemes like overcurrent relays, which are tuned for high-magnitude faults, fail to detect such internal anomalies. Moreover, frequency response deviations caused by TWFs often resemble those introduced by routine phenomena such as tap changer operations, load variation, or core saturation, making accurate diagnosis difficult using traditional FRA interpretation techniques. This paper presents a novel diagnostic framework combining Discrete Wavelet Transform (DWT) and Support Vector Machine (SVM) classification to improve the detection of TWFs. The proposed system employs region-based statistical deviation labeling to enhance interpretability across five well-defined frequency bands. It is validated on five real FRA datasets obtained from operating transformers in Gauteng Province, South Africa, covering a range of MVA ratings and configurations, thereby confirming model transferability. The system supports post-processing but is lightweight enough for near real-time diagnostic use, with average execution time under 12 s per case on standard hardware. A custom graphical user interface (GUI), developed in MATLAB R2022a, automates the diagnostic workflow—including region identification, wavelet-based decomposition visualization, and PDF report generation. The complete framework is released as an open-access toolbox for transformer condition monitoring and predictive maintenance. Full article