Fault Identification Method Research Articles

Research on ultrasonic defect identification method of well control manifold pipeline based on IAFSA-SVM

In view of the classification of corrosion defects of well controlled manifold pipelines, an ultrasonic defect recognition method based on the combination of support vector machine(SVM) and improved artificial fish swarm algorithm (IAFSA) is proposed. Firstly, perform wavelet packet decomposition on the ultrasonic defect signal waveform to obtain the characteristic vector of characterizes the defect type; Then establish the support vector machine defect classification model, and use the improved artificial fish swarm algorithm to optimize the support vector machine parameters. Finally, a software and hardware experimental platform for the classification of pipeline corrosion defects of the well control manifold is built to carry out software simulation and experimental analysis. The experimental results show that the recognition rate of the defect classification model based on improved artificial fish swarm optimization support vector machine parameters is 94.67% for ultrasonic defect signals at different depths.

Identification Methods for Transformer Turn-to-Turn Faults and Inrush Current Based on Electrical Information

Power transformer is the most imperative and crucial equipment in electrical power system. Accurately monitoring the occurrence of transformer faults and rapidly isolating the faults are of great significance to ensure the reliability of power supply and avoid serious damage to the transformer. There are many methods used in the transformer to monitor transformer running state. This paper provides a comprehensive overview of transformer turn-to-turn fault and inrush current identification methods based on electrical information, and expounds the systematic research content that needs to be carried out for transformer turn-to-turn fault identification methods based on magnetic flux leakage information.

Composite Insulator Defect Identification and Quantitative Method Based on Random Hough Transform Ellipse Detection

To reduce the great influence on the operation of the transmission lines resulted from the external force damage of the insulator, and to improve the operation and maintenance level of the lines, a composite insulator defect identification method based on random Hough transform ellipse detection is proposed in this paper. This method takes the image of the insulator taken by drone aerial photography as the research object. First, the edge of the insulator is extracted by the Canny operator. Then, an ellipse curve fitting is applied to the extracted edge. Finally, the edge curve of the composite insulator is compared with the fitted elliptic curve, and the defect area is identified. This method cannot only identify the breakage state of the insulator, but also evaluate the degree of damage by comparing the damaged area.

Research on fault identification method based on multi-resolution permutation entropy and ABC-SVM

Research on fault identification method based on multi-resolution permutation entropy and ABC-SVM

Quantitative analysis of wall thinning of bimetallic clad steel tube based on pulsed eddy current

AbstractBimetallic clad steel tube (BCST) has been widely adopted in onshore and offshore oil and gas transportation due to its excellent corrosion resistance and mechanical properties. Aged BCSTs are susceptible to the initiation of various types of defects, including corrosion, dents, cracking/leakage, and wall‐thinning defects. However, due to the complex structure of BCST and the different materials of inner and outer tubes, it is difficult to implement conventional nondestructive testing. In light of this, a finite element simulation model is established based on the pulsed eddy current testing technology in this paper. Through a series of simulation analyses, the influence of tubing material, defect location, and other factors on the detection results is intensively investigated. By establishing the mapping relationship between the detection signal characteristics and the wall thickness reduction ratio, it is confirmed that there is a quadratic relationship between the differential voltage peak value and the wall thickness reduction, and the defect identification and quantitative evaluation are carried out. The validity of the proposed defect identification and quantitative evaluation method is verified by experiments, which provides a theoretical basis for BCST nondestructive testing and state maintenance. This work facilitates BCSTs structural integrity management and state maintenance decisions, which are critical to the safe operation of the oil and gas system.

Automatic Defect Identification Method for Magnetic Particle Inspection of Bearing Rings Based on Visual Characteristics and High-Level Features

Fluorescent magnetic particle inspection (MPI) is a conventional non-destructive testing process for railway bearing rings that still needs to be completed manually. Due to the complexity of bearing ring surfaces in inspection, automatic detection for bearing rings based on image processing is difficult to apply. Therefore, we proposed a bearing ring defect identification method based on visual characteristics and high-level features. Inspired by the mechanism of human visual perception, defects can be identified from the complex background conveniently by human eyes. According to the linear structure characteristics and greyscale distribution characteristics of cracks in the acquired images, we introduce the centerline extraction and Gaussian similarity measure to reduce background noise and obtain the crack candidate regions. Then, an improved MobileNetV3 is used to extract high-level features of the candidate regions and determine whether they are defective, which uses a new attention module, Coordinate Attention (CA), to substitute the Squeeze-and-Excitation (SE) attention to improve the performance. The experimental results show that the detection accuracy rate of the proposed method is 96.5%. Compared with traditional methods, the proposed method can efficiently extract crack defects in a complex textured background and shows high-quality performance in recall and precision.

Series Arc Fault Identification Method Based on Multi-Feature Fusion

With the increase of various loads connected to the low-voltage distribution system, the difficulty of identifying low-voltage series fault arcs has greatly increased, which seriously threatens the electricity safety. Aiming at such problems, a neural network algorithm based on multi-feature fusion is proposed. The fault current has the characteristics of randomness, high frequency noise, and singularity. A GA-BP neural network model is built, and the wavelet analysis method (based on singularity), Fourier transform method (based on high frequency noise), current cycle difference method (based on randomness), and current cycle similarity derivation method (based on randomness) are used for feature extraction and can more comprehensively reflect the characteristics of arc faults. Simulation results show that the multi-feature fusion algorithm has a higher recognition rate than other algorithms. Moreover, compared with the support vector machine model, logistic regression model, and AlexNet model, the GA-BP neural network model has a higher recognition accuracy than the other three models, which can reach 99%.

Research on a Multisource Domain Improved Fault Diagnosis Method of the Rotor System

In this paper, rotor systems of the rotating machinery such as steam turbines, centrifugal compressors and flue gas turbines are selected as the research objects. At present, most of the rotor system fault diagnosis methods based on artificial intelligence algorithms are in the laboratory research stage, and there is still a gap from the actual industrial application. Therefore, the multi-source domain improved fault diagnosis (MSDIFD) method for satisfying engineering applications is proposed in this paper. Firstly, typical labeled data are selected to construct a multi-source domain training feature space. Then, commonality fault features are extracted and screened by the improved adaptive variational mode decomposition (IAVMD), and the feature reconstruction signal is automatically output. Next, the reinforced semi-supervised transfer component analysis (RSSTCA) method based on enhanced kernel function is employed to narrow the disparity between feature vectors of cross domain data. Finally, typical failure case data and real-time monitoring data are used as the training data and test data of the model, respectively, and an ensemble fault recognition classifier is constructed to achieve failure mode identification of the rotor system. Using 40 groups of typical fault engineering cases under different equipment and different operating conditions, the proposed rotor fault identification method has been verified and compared with five published fault identification methods. The results indicate that the proposed method possesses more excellent fault diagnosis accuracy and domain generalization performance, and the MSDIFD method has good application and promotion value for solving cross equipment, cross working condition, and cross domain diagnostic tasks.

Reliable Algebraic Fault Detection and Identification of Robots

This paper presents a reliable algebraic method for fault detection and identification for rigid robots considering additive faults and external forces. The presented method is based on the reconstruction of a residual or the fault by nonlinear differential algebraic expressions using the polynomial approximation operator. Furthermore, it is shown how the filter parameters and the noise suppression properties of the resulting filter can be parameterized for reliable fault diagnosis according to the characteristics of the fault signals in the frequency domain. The presented method provides an evaluable expression that can be used to explicitly estimate the influence of measurement noise, quantization, numerical errors, and parameter uncertainties in terms of computable error bounds, thus ensuring reliable fault diagnosis. Based on a SCARA robot, a comparison of the method with the generalized momentum observer is performed, highlighting the advantages of the presented approach. Furthermore, it is applied to a real-world application of a delta robot with parallel kinematics for the identification of additive faults while considering interactions due to the application. For this purpose, the appropriate parameterization of the filter is determined and then the feasibility of the method is verified by the implementation on a real-time controller.

Software Defect Fault Intelligent Location and Identification Method Based on Data Mining

With the advancement of the times, computer technology is also constantly improving, and people’s requirements for software functions are also constantly improving, and as software functions become more and more complex, developers are technically limited and teamwork is not tacitly coordinated. And so on, so in the software development process, some errors and problems will inevitably lead to software defects. The purpose of this paper is to study the intelligent location and identification methods of software defects based on data mining. This article first studies the domestic and foreign software defect fault intelligent location technology, analyzes the shortcomings of traditional software defect detection and fault detection, then introduces data mining technology in detail, and finally conducts in-depth research on software defect prediction technology. Through in-depth research on several technologies, it reduces the accidents of software equipment and delays its service life. According to the experiments in this article, the software defect location proposed in this article uses two methods to compare. The first error set is used as a unit to measure the subsequent error set software error location cost. The first error set 1F contains 19 A manually injected error program, and the average positioning cost obtained is 3.75%.

Robustification of Unscented Kalman Filtering to Identify Faults in Gas Turbine Engine

A methodology is proposed to improve the robustification of the Unscented Kalman Filtering by adjusting the measurement noise covariance matrix by introducing two adaptive factors. The combined effect of the two adaptive factors helps to identify and filter the outliers more prominently. Improvement in the proposed robust UKF algorithm is established by analyzing the percentage reduction of accumulation error for different conditions with the larger measurement noise covariance. In addition, a method of fault identification of gas turbine engines is developed by the proposed robust UKF approach in the presence of system and measurement noises. The proposed fault identification method deals with the analysis of the residual signature of different states along with the cost function of the derived algorithm. Different component and sensor anomalies are introduced in the nonlinear dynamic model of the gas turbine engine to produce the faulty situations. The deviations in performances are compared with the nominal behavior to obtain the residual signature. A comparative study is performed to identify the faulty scenarios.

Fault Identification Method Based on Unified Inverse-Time Characteristic Equation for Distribution Network

Fault Identification Method Based on Unified Inverse-Time Characteristic Equation for Distribution Network

Bearing Fault Diagnosis Method of Wind Turbine Based on Improved Anti-Noise Residual Shrinkage Network

Aiming at the difficulty of rolling bearing fault diagnosis of wind turbine under noise environment, a new bearing fault identification method based on the Improved Anti-noise Residual Shrinkage Network (IADRSN) is proposed. Firstly, the vibration signals of wind turbine rolling bearings were preprocessed to obtain data samples divided into training and test sets. Then, a bearing fault diagnosis model based on the improved anti-noise residual shrinkage network was established. To improve the ability of fault feature extraction of the model, the convolution layer in the deep residual shrinkage network was replaced with a Dense-Net layer. To further improve the anti-noise ability of the model, the first layer of the model was set as the Drop-block layer. Finally, the labeled data samples were used for training model and the trained model was applied to the test set to output the fault diagnosis results. The results showed that the proposed method could achieve the fault diagnosis of wind turbine bearing more accurately in the high noise environment through comparison and verification.

A Novel Fault Identification Method Driven by Knowledge and Data

In the field of intelligent manufacturing, fault identification is an effective way to improve product service by identifying the cause of failures. For addressing it, the Generalized Bayesian Network (GBN) model is extended based on the traditional Bayesian Network in this paper, which redefines the directed edges and probability parameters among nodes. Compared with Bayesian Network, the GBN model has the ability to simultaneously define causality and correlation of variables. In addition, the structure of network is not only based on statistical data but also driven by expert knowledge. In order to achieve the collaboration of data and knowledge while maintaining the consistency, a hierarchical collaborative framework is designed including the data layer and knowledge layer. Furthermore, a hierarchical multi-objective optimization algorithm, namely Hierarchical Non-dominated Sorting Genetic Algorithm II (HNSGA-II), is advanced to solve the proposed model. Finally, an industrial case study for fault cause identification targeting the product service helps illustrate all details.

Vehicle Engine Fault Identification Method Equipment and System Based on Deep Learning

In the actual driving process of the vehicle, the engine may have a variety of faults. Timely identification or detection of engine faults is an urgent technical problem to be solved. In the existing traditional technology, engineers with rich experience are usually employed to manually detect engine faults. This method is inefficient and has high error rate. This paper presents an engine fault recognition method, equipment and medium based on deep learning to realize the automatic recognition of engine faults.

A Novel Fault Identification Method for HVDC Converter Station Section Based on Energy Relative Entropy

The ac line fault could lead to the operation of converter station protection for high voltage dc (HVDC) systems. In order to avoid the false operation of converter station protection, a novel fault section identification scheme based on energy relative entropy is proposed. The fault characteristics of internal and external for converter station section are analyzed, and it is found that the difference of forward current traveling wave and backward current traveling wave is obvious when the fault occurs in different section. The difference between the forward and backward current traveling wave is represented by the S-transform energy relative entropy. The S-transform energy relative entropy for the external ac line fault is smaller than that of the internal fault of converter station. The performance and effectiveness of the proposed section identification method are evaluated and verified based on simulation and experiment. Finally, the adaptation is studied by considering the fault position, fault type, fault initial angle, transition resistance, and noise. The proposed method can accurately identify the fault of ac system from the converter station and effectively prevent the converter station protection from incorrect operation.

Fault identification of double-circuit transmission lines on the same pole based on EEMD energy ratio

In order to improve the sensitivity and reliability of traveling wave protection, on the basis of analyzing the relationship of the anti-traveling wave current amplitude in the window after the internal/external failure of the double circuit line on the same tower, a fault identification method based on EEMD energy ratio is proposed. Use EEMD decomposition to decompose the anti-traveling wave current in a time window after the fault into 7 scales, and extracts the EEMD energy ratio at each scale at both ends to form a feature vector. Then it is sent to the particle swarm optimization support vector machine (PSO-SVM) for training and testing, and the internal and external faults are identified. Experiments show that the algorithm has good fault identification ability, the fault accuracy is 95% and the method sensitivity is high.

Wind Turbine Fault Diagnosis by the Approach of SCADA Alarms Analysis

Wind farm operators are overwhelmed by a large amount of supervisory control and data acquisition (SCADA) alarms when faults occur. This paper presents an online root fault identification method for SCADA alarms to assist operators in wind turbine fault diagnosis. The proposed method is based on the similarity analysis between an unknown alarm vector and the feature vectors of known faults. The alarm vector is obtained from segmented alarm lists, which are filtered and simplified. The feature vector, which is a unique signature representing the occurrence of a fault, is extracted from the alarm lists belonging to the same fault. To mine the coupling correspondence between alarms and faults, we define the weights of the alarms in each fault. The similarities is measured by the weighted Euclidean distance and the weighted Hamming distance, respectively. One year of SCADA alarms and maintenance records are used to verify the proposed method. The results show that the performance of the weighted Hamming distance is better than that of the weighted Euclidean distance; 84.1% of alarm lists are labeled with the right root fault.

Fault identification method for distribution network based on parameter optimized variational mode decomposition and convolutional neural network

In order to extract the fault characteristics of the distribution power system automatically, a new method for fault identification based on parameter optimized variational mode decomposition (VMD) and convolutional neural network (CNN) is proposed. First, the VMD is used to decompose electric signals of the main transformer, including the current of the low-voltage inlet line, bus voltage, and so on. The obtained intrinsic mode functions should be arranged in sequence from low to high according to the central frequency. The VMD can obtain the time-frequency matrices. In order to decide the decomposition layers of the VMD adaptively, the partial mean of multi-scale permutation entropy is used to optimize it. Then, the time-frequency matrixes can be converted into pixel matrices by pseudo-colour coding. The pixel matrices which are appropriately compressed can be used as input of CNN. The CNN can autonomously extract the eigenvectors of fault. Finally, the classification algorithms of support vector machine (SVM), naive Bayesian classifier (NBC), extreme learning machine (ELM), and random forest (RF) are used to train and test the feature vectors extracted by CNN to verify the effectiveness of the method. The experiments show that the proposed method has excellent stability, fast convergence speed, and high precision. The technique can effectively complete the fault identification of the distribution network.

VMD based trigonometric entropy measure: a simple and effective tool for dynamic degradation monitoring of rolling element bearing

Early identification of rolling element defects is always a topic of interest for researchers and the industry. For early fault identification, a simple and effective dynamic degradation monitoring method using variational mode decomposition (VMD) based trigonometric entropy measure is developed. First, vibration signals are obtained and are further decomposed using VMD to obtain various frequency modes. Second, a trigonometric entropy measure is developed to monitor the dynamic change occurring in the health of bearing. Third, trigonometric entropy measure of various VMD modes is computed. Fourth, the variance of measure is computed and two modes having the highest variance are selected for principal component analysis (PCA). Thereafter, PCA of selected measures is carried out. Finally, dynamic degradation monitoring is carried out by observing the trend in the principal component having the highest diverse information. The testing of newly developed VMD based trigonometric entropy measure is carried out on the two different types of data set. One is from XJTU-SY Bearing datasets and another is from the Centre for Intelligent Maintenance Systems. The experimental study reveals that the proposed method is capable of raising the alarm about the initiation of defects at a very early stage. Compared to existing indicators such as kurtosis, RMS, and Shannon entropy, the proposed method is superior while carrying out defect degradation monitoring.