Efficiency Of Fault Diagnosis Research Articles

Refined composite hierarchical bidirectional diversity entropy and its application in fault type diagnosis and fault degree diagnosis of diesel engine fuel injectors

The performance of a diesel engine is closely related to the injector, and the fuel pressure shock signal for monitoring the state of the injector always contains multiple natural oscillation modes at different levels. Therefore, it is very necessary to detect the dynamic changes of the injector signal from multiple levels. Multi-scale Diversity Entropy (MDE) is a recent commonly used method of information entropy. However, due to its inherent defects, it limits the wide application of MDE in fault feature extraction. Therefore, this paper proposes a new nonlinear dynamics method, called the Refined Composite Hierarchical Bidirectional Diversity Entropy (RCHBDE). The ability of RCHBDE to extract signal complexity changes was verified through simulation signals, which is superior to MDE and multi-scale bidirectional diversity entropy (MBDE). On this basis, this article proposes a diagnostic framework for fuel injector fault type and degree based on RCHBDE and Seagull optimized Support Vector Machine. We apply the proposed method to the injector failure test data. The results show that RCHBDE can better reflect the dynamic change process of the injector fault signal than MDE and MBDE. Moreover, compared with existing methods, the proposed method in this paper improves the accuracy and efficiency of fault diagnosis.

Fault detection of high-voltage insulator thermal images based on YOLOv5

In this paper, we propose a defect detection method based on YOLOv5 for thermal images of high-voltage insulators, in which a large number of insulator photographs are taken by a thermal imager, followed by a YOLOv5 model to detect the captured thermal images. The function Meta-ACON is proposed to replace the original activation function. Finally, the extracted features are used for insulator condition identification and defect detection. Experimental findings demonstrate the efficacy of the suggested approach in successfully accomplishing the feature extraction of insulator faults, thus improving the accuracy and efficiency of insulator fault diagnosis.

A study on the decoupling of helical gear compound uniform wear and tooth root crack faults via gear condition indicators based on dynamic modeling

Diagnosis of compound faults plays a significant role in evaluating the health status of a gearbox. The difficulty of compound fault diagnosis is due to the entanglement of fault symptoms. In this work, diagnosis of single and compound uniform wear and tooth root crack via gear condition indicators have been studied analytically. A three-dimensional dynamic model that can include several design parameters and working conditions has been employed to model the dynamic of compound faults for the first time. The effect of fault severity, noise, working conditions, and design parameters on the diagnosis efficiency have been investigated. The qualitative and quantitative assessment of fault diagnosis has been accomplished via t-SNE and Generalized Discriminative Value (GDV) measures, respectively. The results show that the gear condition indicators are more suitable to assess the general health state of the gearbox. However, their ability to detect the type of fault and decouple compound faults are limited. Furthermore, design parameters of the gear transmission system, such as helix angle and gear width, can highly affect the fault diagnosis efficiency.

Early rolling bearing fault diagnosis in induction motors based on on-rotor sensing vibrations

The traditional on-house sensing (OHS) accelerometer for vibration measurements causes poor signal-to-noise ratio (SNR) and complicated fault modulations, which increases the difficulty and complexity for early bearing fault diagnosis. To overcome these challenges, this paper develops a wireless triaxial on-rotor sensing (ORS) system to largely improve the SNR and deduces fast Fourier transform (FFT) and Hilbert envelope analysis for accurate early rolling bearing fault diagnosis, which largely improves accuracy and efficiency for early fault diagnosis. First, the development of the ORS system for wireless vibration measurements is given. Second, the theoretical diagnostic relationships between dynamic ORS signals and rolling bearing faults are derived for FFT and Hilbert envelope analysis for the first time. Finally, the induction motor tests with outer and inner race faults successfully validate that both simple FFT and Hilbert envelope analysis can achieve more robust early rolling bearing fault diagnosis compared to OHS measurements.

LogBASA: Log Anomaly Detection Based on System Behavior Analysis and Global Semantic Awareness

System log anomaly detection is important for ensuring stable system operation and achieving rapid fault diagnosis. System log sequences include data on the execution paths and time stamps of system tasks in addition to a large amount of semantic information, which enhances the reliability and effectiveness of anomaly detection. At the same time, considering the correlation between system log sequences can effectively improve fault diagnosis efficiency. However, the existing system log anomaly detection methods mostly consider only the sequence patterns or semantic information on the logs, so their anomaly detection results show a high rate of missed and false alarms. To solve these problems, this paper proposed an unsupervised log anomaly detection model (LogBASA) based on the system behavior analysis and global semantic awareness, aiming to decrease the leakage rate and increase the log sequence anomaly detection accuracy. First, a system log knowledge graph was constructed based on massive, unstructured, and multilevel system log data to represent log sequence patterns, which facilitates subsequent anomaly detection and localization. Then, a self-attention encoder-decoder transformer model was developed for log spatiotemporal association analysis. This model combines semantic mapping and spatiotemporal features of log sequences to analyze system behavior and log semantics in multiple dimensions. Furthermore, a system log anomaly detection method that combines adaptive spatial boundary delineation and sequence reconstruction objective functions was proposed. This method uses special words to characterize the log sequence states, delineates anomaly boundaries automatically, and reconstructs log sequences through unsupervised training for anomaly detection. Finally, the proposed method was verified by numerous experiments on three real datasets. The results indicate that the proposed method can achieve an accuracy rate of 99.3%, 95.1%, and 97.2% on HDFS, BGL, and Thunderbird datasets, which proves the effectiveness and superiority of the LogBASA model.

Diagnosis of single and multiple-source faults of chiller sensors using EWEEMD-ICKNN by time sequence denoising and non-Gaussian distribution feature extraction

Currently, fault detection of chiller sensors primarily focuses on single-source fault (SSF) sensors. These methods are typically ineffective in multiple source faults (MSF) cases. These methods assume a Gaussian distribution of the chiller data and ignore the coupling between the sensors. Therefore, we propose temporal feature denoising and non-Gaussian distribution-based methods in this study. First, the chiller sensor signals are denoised based on exponential weighted moving average ensemble empirical mode decomposition (EWEEMD). Subsequently, an independent component analysis k-nearest neighbour (ICKNN) classifier is proposed based on the non-Gaussian distribution features of the chiller sensor data. A residual vector combined with independent component analysis was used for feature extraction to propose the ICKNN classifier. Finally, according to their characteristics, different SSF and MSF were applied to the chiller sensors. Experimental comparisons with other methods were performed for four single-source faults and six multisource faults. The EWEEMD-ICKNN was experimentally confirmed to have excellent fault detection performance in the SSF and MSF chiller sensors. The experimental results demonstrate that the fault diagnosis rate of EWEEMD-ICKNN ranges from 99.90% to 86.80%. Additionally, the fault diagnosis efficiency of EWEEMD-ICKNN remains consistently stable across various fault states, exhibiting an average F1 score of 0.9589 with F1 score statistics of 0.038.

Fault Diagnosis of Support Vector Machine Analog Circuits Based on Improved Particle Swarm Optimization

The development of electronic circuits requires that the reliability and security of circuit equipment and system operation are also increasing. In addition, due to the complexity of the operating environment, it is very important to strengthen the fault diagnosis and real-time testing technology of analog circuits in circuit systems. Based on this, this paper studied the fault diagnosis of analog circuits with Support Vector Machine (SVM), and introduced Improved Particle Swarm Optimization (IPSO) algorithm to optimize the parameters of SVM. In other words, the dynamic weight setting and factor improvement of Particle Swarm Optimization (PSO) algorithm aim to accelerate algorithm performance improvement, and information extraction and diagnosis model construction are carried out on the basis of considering circuit fault characteristics. Through the performance test and application analysis of the improved algorithm proposed in the study, the error value of the improved algorithm was basically stable at 0.0103 in the late stage of classification training, and its prediction accuracy rate was more than 80%, and the classification consumption time was less. At the same time, the accuracy of fault feature extraction results in training and test scenarios was above 94%, and the search performance was obviously better than other comparison algorithms, which effectively improved the fault diagnosis accuracy and efficiency. The IPSO algorithm model can effectively identify analog circuit fault information, and shows good information optimization performance. It has certain validity and rationality in circuit fault diagnosis and security assurance.

Hydraulic Systems Fault Diagnosis Based on Random Forests Recursive Feature Elimination and XGBoost

This paper proposes an RFRFE-XGBoost method for fault mode recognition in hydraulic systems. The proposed method combines random forests-based recursive feature elimination (RFRFE) and extreme gradient boosting (XGBoost) to effectively identify important features and improve fault diagnosis efficiency and accuracy. The method is validated on relevant datasets and compared with existing methods, demonstrating its effectiveness and superiority. The results show that RFRFE-XGBoost can accurately recognize various fault modes and outperforms other methods in terms of classification accuracy and computational efficiency. The proposed method provides a promising approach for fault diagnosis in complex systems.

Convolutional-Transformer Model with Long-Range Temporal Dependencies for Bearing Fault Diagnosis Using Vibration Signals

Fault diagnosis of bearings in rotating machinery is a critical task. Vibration signals are a valuable source of information, but they can be complex and noisy. A transformer model can capture distant relationships, which makes it a promising solution for fault diagnosis. However, its application in this field has been limited. This study aims to contribute to this growing area of research by proposing a novel deep-learning architecture that combines the strengths of CNNs and transformer models for effective fault diagnosis in rotating machinery. Thus, it captures both local and long-range temporal dependencies in the vibration signals. The architecture starts with CNN-based feature extraction, followed by temporal relationship modelling using the transformer. The transformed features are used for classification. Experimental evaluations are conducted on two datasets with six and ten health conditions. In both case studies, the proposed model achieves high accuracy, precision, recall, F1-score, and specificity all above 99% using different training dataset sizes. The results demonstrate the effectiveness of the proposed method in diagnosing bearing faults. The convolutional-transformer model proves to be a promising approach for bearing fault diagnosis. The method shows great potential for improving the accuracy and efficiency of fault diagnosis in rotating machinery.

Surface defects detection in metal materials repaired by laser surfacing of seal welds

Laser surfacing repair technology for sealing welds is widely used in metal repair. Due to welding technology and usage scenarios, process defects on the metal surface are inevitable. Therefore, ultrasonic surface wave technology is used to analyze the surface defects of metal materials. Principal Component Analysis (PCA) is used to extract the main defect signals on the metal surface, and synthetic aperture focusing technology is used to reduce imaging errors. Considering the lack of PCA in imaging defects, wavelet domain hidden Markov models (WHMM) are combined to optimize the signal, thereby improving the inspection effect of metal defects. In the test results of the relationship between the propagation distance of 316 L steel and the defect echo signal, the echo signal gradually fitted as the propagation distance increased. When the propagation distance was greater than 10 mm, the image acquisition defect signal had significant noise points. Various techniques were used to process the original echo signals of metal surface defects. The improved PCA-WHMM algorithm had significant advantages with the SNR value of the defect image increased by 13.65 % compared to PCA-WHMM. At the same time, the surface repair effects of laser surfacing 316 L metal before and after optimization were compared. The hardness, toughness, and corrosion resistance of the optimized metal were significantly improved. The proposed technological innovation combines traditional laser surfacing repair with deep learning fault diagnosis, which not only greatly improves the efficiency of fault diagnosis, but also proves that this research can effectively avoid common focus issues of laser surfacing repair technology, providing important technical reference for the application of ultrasonic technology in metal defect detection.

Bearing Fault Diagnosis Method Based on Deep Learning and Health State Division

As a key component of motion support, the rolling bearing is currently a popular research topic for accurate diagnosis of bearing faults and prediction of remaining bearing life. However, most existing methods still have difficulties in learning representative features from the raw data. In this paper, the Xi’an Jiaotong University (XJTU-SY) rolling bearing dataset is taken as the research object, and a deep learning technique is applied to carry out the bearing fault diagnosis research. The root mean square (RMS), kurtosis, and sum of frequency energy per unit acquisition period of the short-time Fourier transform are used as health factor indicators to divide the whole life cycle of bearings into two phases: the health phase and the fault phase. This division not only expands the bearing dataset but also improves the fault diagnosis efficiency. The Deep Convolutional Neural Networks with Wide First-layer Kernels (WDCNN) network model is improved by introducing multi-scale large convolutional kernels and Gate Recurrent Unit (GRU) networks. The bearing signals with classified health states are trained and tested, and the training and testing process is visualized, then finally the experimental validation is performed for four failure locations in the dataset. The experimental results show that the proposed network model has excellent fault diagnosis and noise immunity, and can achieve the diagnosis of bearing faults under complex working conditions, with greater diagnostic accuracy and efficiency.

Exploring Research on the Construction and Application of Knowledge Graphs for Aircraft Fault Diagnosis.

Fault diagnosis is crucial for repairing aircraft and ensuring their proper functioning. However, with the higher complexity of aircraft, some traditional diagnosis methods that rely on experience are becoming less effective. Therefore, this paper explores the construction and application of an aircraft fault knowledge graph to improve the efficiency of fault diagnosis for maintenance engineers. Firstly, this paper analyzes the knowledge elements required for aircraft fault diagnosis, and defines a schema layer of a fault knowledge graph. Secondly, with deep learning as the main method and heuristic rules as the auxiliary method, fault knowledge is extracted from structured and unstructured fault data, and a fault knowledge graph for a certain type of craft is constructed. Finally, a fault question-answering system based on a fault knowledge graph was developed, which can accurately answer questions from maintenance engineers. The practical implementation of our proposed methodology highlights how knowledge graphs provide an effective means of managing aircraft fault knowledge, ultimately assisting engineers in identifying fault roots accurately and quickly.

New criteria for wrapper feature selection to enhance bearing fault classification

Classification is a critical task in many fields, including signal processing and data analysis. The accuracy and stability of classification results can be improved by selecting the most relevant features from the data. In this paper, a new criterion for feature selection using wrapper method is proposed, which is based on the evaluation of the classification results according to the accuracy and stability (standard deviation) of each class and the number of selected features. The proposed method is evaluated using Random Forest (RF) and Ant Colony Optimization (ACO) algorithms on a benchmark dataset. Results show that the proposed method outperforms classical feature selection methods in terms of accuracy and stability of classification results, especially for the difficult-to-classify combined damage class. This study demonstrates the effectiveness of the proposed new wrapper feature selection criterion to improve the performance of classification algorithms with higher stability (STD: C1 = 0.5, C2 = 0.8, C3 = 0.6, C4 = 1.8) and better accuracy (average C1 = 98.5%, C2 = 96.6%, C3 = 9.5%, C4 = 93) for the both; the statoric current and the vibration signal compared to other techniques. Machine learning methods had proven their efficiency in time-varying machines fault diagnosis when taking vibration signals and statoric currents extracted features as inputs. However, the use of the both demonstrated a higher robustness and a remarkable superiority.

An effective approach based on nonlinear spectrum and improved convolution neural network for analog circuit fault diagnosis.

In this work, an effective approach based on a nonlinear output frequency response function (NOFRF) and improved convolution neural network is proposed for analog circuit fault diagnosis. First, the NOFRF spectra, rather than the output of the system, are adopted as the fault information of the analog circuit. Furthermore, to further improve the accuracy and efficiency of analog circuit fault diagnosis, the batch normalization layer and the convolutional block attention module (CBAM) are introduced into the convolution neural network (CNN) to propose a CBAM-CNN, which can automatically extract the fault features from NOFRF spectra, to realize the accurate diagnosis of the analog circuit. The fault diagnosis experiments are carried out on the simulated circuit of Sallen-Key. The results demonstrate that the proposed method can not only improve the accuracy of analog circuit fault diagnosis, but also has strong anti-noise ability.

Multivariate local fluctuation mode decomposition and its application to gear fault diagnosis

In this paper, we propose a novel method, called multivariate local fluctuation mode decomposition (MLFMD), to improve the accuracy and efficiency of fault diagnosis using multiple channels signals. Compared with multivariate empirical mode decomposition (MEMD), MLFMD uses second-order differentiable local extreme point localization (SDLEPL) to mine the local hidden information and an adaptive non-uniform projection (ANP) technique to improve the decomposition accuracy. In addition, the MLFMD method employs multivariate periodic mean to extract the mean curves, which improves the decomposition efficiency. Compared with traditional MEMD, our proposed MLFMD algorithm has higher decomposition accuracy and efficiency. Furthermore, a new fault diagnosis method based on MLFMD is proposed, which can efficiently fuse data from each channel. The efficacy of the proposed method is validated with both simulated and real-world signals, and the results demonstrate the superiority of the MLFMD.

A hybrid intelligent gearbox fault diagnosis method based on EWCEEMD and whale optimization algorithm-optimized SVM

PurposeThis paper proposes an intelligent fault diagnosis method, which aims to obtain the outstanding fault diagnosis results of the gearbox.Design/methodology/approachAn intelligent fault diagnosis method based on energy entropy-weighted complementary ensemble empirical mode decomposition (EWCEEMD) and support vector machine (SVM) optimized by whale optimization algorithm (WOA) is proposed. The raw signal is first denoised by the wavelet noise reduction method. Then, complementary ensemble empirical mode decomposition (CEEMD) is used to generate several intrinsic mode functions (IMFs). Next, energy entropy is used as an indicator to measure the sensibility of the IMF and converted into a weight coefficient by function. After that, IMFs are linearly weighted to form the reconstruction signal, and several features are extracted from the new signal. Finally, the support vector machine optimized by the whale optimization algorithm (WOA-SVM) model is used for gearbox fault classification using feature vectors.FindingsThe fault features extracted by this method have a better clustering effect and clear boundaries under each fault mode than the unimproved method. At the same time, the accuracy of fault diagnosis is greatly improved.Originality/valueIn most studies of fault diagnosis, the sensitivity of IMF has not been appreciated. In this paper, energy entropy is chosen to quantify sensitivity. In addition, high classification accuracy can be achieved by applying WOA-SVM as the final classification model, improving the efficiency of fault diagnosis as well.

Transformer online monitoring and intelligent diagnosis system design

Transformer is a pivotal equipment in the power system, and its safe and stable operation affects the entire power system. Therefore, the on-line monitoring and fault diagnosis of transformers are particularly important in practical applications. The traditional maintenance method is called regular maintenance. This maintenance method is to call a large number of inspection workers at regular intervals to carry out large-scale inspections, not only to collect various working data of the transformer, but also to observe its working status, and finally manually. Check whether there are hidden dangers in major machines, and judge whether the transformer can run safely and stably according to the data obtained by these large amounts of manpower and material resources. In order to improve the efficiency and accuracy of transformer fault diagnosis, and to carry out all-round real-time online monitoring and fault diagnosis of 220kV power transformers, considering the actual needs of electricity companies, this paper is based on the Qt development platform, C++, MySQL and other development environments. And developed a transformer online monitoring and intelligent diagnosis system, which mainly has the functions of providing basic information management, online monitoring, fault diagnosis, giving operation and maintenance suggestions, and providing users with visual comprehensive analysis. Tested with actual data, it is proved that the system can objectively reflect the operating state of the transformer and carry out real-time fault diagnosis. The system designed in this paper meets the needs of operators to understand the status of distribution transformers, and effectively solves the problem of untimely detection of distribution transformer faults. It can effectively improve the efficiency of detection, reduce the probability of major accidents, and reduce operation and maintenance costs., which saves financial and human resources, and is of great significance for improving economic efficiency and personal and equipment safety.

Implementation of Hybrid Particle Swarm Optimization for Optimized Regression Testing

Software test case optimization improves the efficiency of the software by proper structure and reduces the fault in the software. The existing research applies various optimization methods such as Genetic Algorithm, Crow Search Algorithm, Ant Colony Optimization, etc., for test case optimization. The existing methods have limitations of lower efficiency in fault diagnosis, higher computational time, and high memory requirement. The existing methods have lower efficiency in software test case optimization when the number of test cases is high. This research proposes the Tournament Winner Genetic Algorithm (TW-GA) method to improve the efficiency of software test case optimization. Hospital Information System (HIS) software was used to evaluate TW-GA model performance in test case optimization. The tournament Winner in the proposed method selects the instances with the best fitness values and increases the exploitation of the search to find the optimal solution. The TW-GA method has higher exploitation that helps to find the mutant and equivalent mutation that significantly increases fault diagnosis in the software. The TW-GA method discards the information with a lower fitness value that reduces the computational time and memory requirement. The TW-GA method requires 5.47 s and the MOCSFO method requires 30 s for software test case optimization.

An intelligent fault detection approach for digital integrated circuits through graph neural networks.

To quickly and accurately realize the fault diagnosis of analog circuits, this paper introduces the graph neural network method and proposes a fault diagnosis method for digital integrated circuits. The method filters the signals present in the digital integrated circuit to remove noise signals and redundant signals and analyzes the digital integrated circuit characteristics after the filtering process to obtain the digital integrated circuit leakage current variation. To the problem of the lack of a parametric model for Through-Silicon Via (TSV) defect modeling, the method of TSV defect modeling based on finite element analysis is proposed. The common TSV defects such as voids, open circuits, leakage, and unaligned micro-pads are modeled and analyzed by using industrial-grade FEA tools Q3D and HFSS, and the equivalent circuit model of resistance inductance conductance capacitance (RLGC) for each defect is obtained. Finally, the superior performance of this paper in fault diagnosis accuracy and fault diagnosis efficiency is verified by comparing and analyzing with the traditional graph neural network method and random graph neural network method for active filter circuits.

Fault Diagnosis of Harmonic Drives Based on an SDP-ConvNeXt Joint Methodology

Harmonic Drives (HDs) are key components of industrial robots, and damages or failures of the HDs can lead to robot paralysis or even cause operational accidents. To prevent this from happening, it is of great importance to design proper algorithms for fault diagnosis of the HDs, especially in real HD manufacturing lines. In this paper, a fault diagnosis method was proposed to convert multiple vibration signals into symmetrized dot pattern (SDP) images, and to use a ConvNeXt model with Transformer training procedures to classify fault features of the HDs under various fault states and working conditions. The method was experimentally validated on three datasets with a diagnostic accuracy of up to 98.5% and 95.3% under constant and variable conditions, respectively, both higher over those using either state-of-the-art convolutional neural networks or Transformers. This was the first time where the SDP and the ConvNeXt were unitedly used to give superior visibility, interpretability, accuracy and efficiency for HD fault diagnosis, indicating good feasibility of using the ConvNeXt as a backbone for this task.