State Recognition Method Research Articles

Analysis of fault of insulation aging of oiled paper of a large‐scale power transformer and the prediction of its service life

Insulation aging of oiled paper is a common malfunction of transformers. It is of great value to recognize the insulation aging status of oiled paper and predict the service life of transformers using appropriate methods. In this study, an aging state recognition method based on partial discharge was proposed. Partial discharge signals were collected, the feature values were obtained through analysis of the principle component factor, and the state was identified by using a support vector machine. A Weibull distribution‐based method was proposed for the prediction of service life, and the residual life of transformers was determined using failure rate function. Through analysis, it was found that the fault analysis method had an accuracy rate of 80.55% in the recognition of the aging state, which was higher than that of Hidden Markov Model (HMM). In the case analysis, the accuracy rate of the method in recognizing the aging state of 100 transformers reached 82.06%. The accuracy of the failure rate function was also verified in the prediction of service life. This study provides some new ideas for the effective recognition of insulation aging of oiled paper in transformers and the prediction of service life. © 2019 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

A novel aging state recognition method of a viscoelastic sandwich structure based on permutation entropy of dual-tree complex wavelet packet transform and generalized Chebyshev support vector machine

A viscoelastic sandwich structure is widely used in mechanical equipment, but therein viscoelastic layers inevitably suffer from aging which changes the dynamic characteristics of the structure and influences the whole performance of the equipment. Hence, accurate and automatic aging state recognition of the viscoelastic sandwich structure is very significant to monitor structural health state and guarantee equipment operating reliably. To fulfill this task, by analyzing the sensor-based vibration response signals, a novel aging state recognition approach of the viscoelastic sandwich structure based on permutation entropy of dual-tree complex wavelet packet transform and generalized Chebyshev support vector machine is proposed in this article. To extract effective aging feature information, the measured nonlinear and non-stationary vibration response signals are processed by dual-tree complex wavelet packet transform, and multiple permutation entropy features are extracted from the frequency-band signals to reflect structural aging states. For accurate and automatic aging state classification, generalized Chebyshev kernel is introduced, and multi-class generalized Chebyshev support vector machine is developed to classify structural aging states. In order to demonstrate the effectiveness of the proposed method, a typical viscoelastic sandwich structure is designed and fabricated, and various structural aging states are created through the hot oxygen–accelerated aging of viscoelastic layers. The testing results show that the proposed method can recognize the different structural aging states accurately and automatically. In addition, the superiority of dual-tree complex wavelet packet transform in processing vibration response signals and the performance of generalized Chebyshev support vector machine in classifying structural aging states are respectively validated by comparing with the commonly used methods.

Rapid Transformer Health State Recognition Through Canopy Cluster-Merging of Dissolved Gas Data in High-Dimensional Space

Dissolved gases in oil are major parameters for assessing the health state of power transformers. Recognizing that the existing state recognition methods are often restrained by data fluctuation and usually require greater computational load, this paper proposes a rapid transformer health state recognition method through Canopy cluster-merging of dissolved gas data in high-dimensional space. Following the introduction of fluctuation coefficient to evaluate the quality of data and the assignation of weight to reflect the difference between gases, the variable-weighted high-dimensional space of dissolved gases is established to suppress the impact from the fluctuated data. The novel Canopy cluster-merging method that overcomes the instability and high computational complexity of the conventional clustering method is then proposed and used in the variable-weighted high-dimensional space to recognize the abnormal state transformer. Applying the state recognition rules and matching with the established abnormal event base, the health state of the transformer could be rapidly recognized. The single case verification concludes that the proposed method has better clustering effect and can significantly improve the clustering speed even by 17.08 times. The group verification test indicates that the proposed method not only demonstrates an accuracy as high as 91.43% but also shows extremely high efficiency compared with the conventional recognition methods.

Study on Planetary Gear Degradation State Recognition Method Based on the Features With Multiple Perspectives and LLTSA

Planetary gear is an important part of the transmission system of large electromechanical equipment. Therefore, it is very important to monitor the degradation of the state of the planetary gear. A method for the degradation state recognition of planetary gear based on the features with multiple perspectives and linear local tangent space alignment (LLTSA) algorithm is presented. First, the time domain features of the original vibration signal are extracted, which have the statistical properties and global significance. Then, the detailed features which pay more attention to the detailed information of the vibration signal are extracted on the basis of improved complete ensemble empirical mode decomposition with adaptive noise, and all those features constitute high dimensional original features. In order to solve the problems of information redundancy and interference features, the original features are processed by LLTSA, and the extraction of low dimensional sensitive features can be achieved. Finally, the optimized support vector machine is studied to recognize the low dimensional sensitive features. The result shows that the proposed method can recognize different degradation states of planetary gear accurately and effectively.

Research on Degradation State Recognition of Planetary Gear Based on Multiscale Information Dimension of SSD and CNN

Planetary gear is the key part of the transmission system for large complex electromechanical equipment, and in general, a series of degradation states are undergone and evolved into a local fatal fault in its full life cycle. So it is of great significance to recognize the degradation state of planetary gear for the purpose of maintenance repair, predicting development trend, and avoiding sudden fault. This paper proposed a degradation state recognition method of planetary gear based on multiscale information dimension of singular spectrum decomposition (SSD) and convolutional neural network (CNN). SSD can automatically realize the embedding dimension selection and component grouping segmentation, and the original vibration signal being nonlinear and nonstationary can be decomposed into a series of singular spectrum decomposition components (SSDCs), adaptively. Then, the multiscale information dimension which combines multiscale analysis and fractal information dimension is proposed for quantifying and extracting the feature information contained in each SSDC. Finally, CNN is used to achieve the effective recognition of the degradation state of planetary gear. The experimental results show that the proposed method can accurately recognize the degradation state of planetary gear, and the overall recognition rate is up to 97.2%, of which the recognition rate of normal planetary gear reaches 100%.

Phenomenon analysis and state classification of surface discharge on oil-impregnated pressboard under AC-DC combined voltage

Surface discharge may cause irreversible damage to turn-to-ground insulation in valve windings of converter transformer, where withstand with AC-DC combined voltage. This paper analyzes the phenomenon and characteristics of surface discharge on oil-impregnated pressboard (OIP) under AC-DC combined voltage, and develops a discharge state recognition method. The cylinder-plate discharge model was used to simulate surface discharge. The results showed that discharge development and OIP failure were significantly accelerated by white marks on OIP which were essentially gaseous channels. The discharge characteristics before and after white mark occurrence were both dominated by AC component because of its bigger contribution to electrical field distribution (EFD), and the DC component had obvious effect on accelerating OIP failure. A set of features representing discharge state after white mark occurrence was selected out by the entropy weight method (EWM), based on which the discharge process was classified into three states (stable, fast development and pre-breakdown state) by fuzzy C means clustering method (FCM). A support vector machine (SVM) classifier to recognize discharge state was trained and showed a good performance, whose average assessment accuracy was up to 91.98%. Moreover, the ratio between negative and positive discharge numbers could be used as an auxiliary indicator of pre-breakdown state.

Hessian-regularized weighted multi-view canonical correlation analysis for working condition recognition of sucker-rod pumping wells

ABSTRACTIn order to more accurately recognize and understand the working condition of sucker-rod pumping wells so as to maximally reduce the cost and increase the profit, a large amount of data has been collected during oil production with sucker-rod pumping wells. In view of the sucker-rod pumping production system in big data and IOT (Internet of things) of oil-gas production, to solve the limitations in the existing working condition recognition research and further improve the recognition accuracy and practicality with fewer labelled working condition samples by utilizing the measured parameters from multiple information sources effectively, in this paper, a novel working condition recognition method based on Hessian-regularized weighted multi-view canonical correlation analysis is proposed. Firstly, the features of the measured ground dynamometer cards, electrical power, wellhead temperature and wellhead pressure data are extracted as four different feature views based on the prior information, empirical knowledge and mechanism analysis. Then a model based on Hessian-regularized weighted multi-view canonical correlation analysis and cosine nearest neighbour multi-classification algorithm is established. The proposed method is applied to the recognition of eleven kinds of working conditions from sixty sucker-rod pumping wells in a certain block in Shengli Oilfield, China. In the case where there are small number of labelled training samples, based on cosine nearest neighbour classification method, the recognition rates are increased by 3.44% and 1.5% compared with traditional recognition methods based on measured ground dynamometer cards and electrical power data, respectively. In contrast to methods based on traditional multi-sources of feature connection, multi-view canonical correlation analysis as well as the unweighted Hessian-regularized multi-view canonical correlation analysis, the recognition rates are increased by 4.46%, 2.21% and 1.62%, respectively.

Deep Denoising Autoencoding Method for Feature Extraction and Recognition of Vehicle Adhesion Status

Feature extraction and classification for deep learning are studied to recognize the problem of vehicle adhesion status. Data concentration acquired by automobile sensors contains considerable noise. Thus, a sparse autoencoder (stacked denoising autoencoder) is introduced to achieve network weight learning, restore original pure signal data by use of overlapping convergence strategy, and construct multiclassification support vector machine (SVM) for classification. The sensors are adopted in different road environments to acquire data signals and recognize the adhesion status online. Results show that the proposed method can achieve higher accuracies than those of the adhesion status recognition method based on SVM and extreme learning machine.

Research on Recognition Method of Driving Fatigue State Based on Sample Entropy and Kernel Principal Component Analysis

In view of the nonlinear characteristics of electroencephalography (EEG) signals collected in the driving fatigue state recognition research and the issue that the recognition accuracy of the driving fatigue state recognition method based on EEG is still unsatisfactory, this paper proposes a driving fatigue recognition method based on sample entropy (SE) and kernel principal component analysis (KPCA), which combines the advantage of the high recognition accuracy of sample entropy and the advantages of KPCA in dimensionality reduction for nonlinear principal components and the strong non-linear processing capability. By using support vector machine (SVM) classifier, the proposed method (called SE_KPCA) is tested on the EEG data, and compared with those based on fuzzy entropy (FE), combination entropy (CE), three kinds of entropies including SE, FE and CE that merged with KPCA. Experiment results show that the method is effective.

Process working condition recognition based on the fusion of morphological and pixel set features of froth for froth flotation

Process condition recognition is an effective way to improve the froth process performance. In previous condition recognition algorithms based on machine vision in flotation process, the used features including gray value, bubble size distribution, load, etc., are essentially statistical results of gray level images and there is local bubble structure information loss in their extraction procedure. Meanwhile, the large number of image data are not adequately utilized. Thus, in this paper, deep neural network are used to extract the pix set features, and a two-step condition recognition method based on bubble image morphology and pixel set features is proposed, which utilizes the large number of image data. First, froth images are segmented into single bubble images. Next, the morphological feature vector of a single bubble image is extracted and classification labels are assigned to the single bubble images via K-means clustering. A large quantity of historical images is analyzed and labeled to train a convolutional neural network (CNN) by which the pixel set features of each bubble image are extracted. The morphological feature vector and pixel set features of the bubble images are then fused for bubble image clustering using the weighted mean-shift algorithm. The frequencies of various types of bubbles in a froth image are calculated to form a bubble frequency set for the froth image. A two-step working condition recognition strategy based on image sequence over a time period is then proposed. In this strategy, the bubble frequency sets of all froth images and those of the bubble images segmented from the froth images over the corresponding time period are matched with those of the images of typical flotation conditions in two main steps to determine the current working condition. Test results using industrial data demonstrate the high accuracy and calculation speed of the proposed method.

Study on HEV’s driving condition recognition method based on PSO algorithm

Study on HEV’s driving condition recognition method based on PSO algorithm

State Recognition of High Voltage Isolation Switch Based on Background Difference and Iterative Search

The automatic recognition of the high voltage isolation switch by remote video monitoring is an effective means to ensure the safety of the personnel and the equipment. The existing methods mainly include two ways: improving monitoring accuracy and adopting target detection technology through equipment transformation. Such a method is often applied to specific scenarios, with limited application scope and high cost. To solve this problem, a high voltage isolation switch state recognition method based on background difference and iterative search is proposed in this paper. The initial position of the switch is detected in real time through the background difference method. When the switch starts to open and close, the target tracking algorithm is used to track the motion trajectory of the switch. The opening and closing state of the switch is determined according to the angle variation of the switch tracking point and the center line. The effectiveness of the method is verified by experiments on different switched video frames of switching states. Compared with the traditional methods, this method is more robust and effective.

Spacecraft Anomaly Recognition Based on Morphological Variational Mode Decomposition and JRD

Considering the difficulty in identifying the in-orbital spacecraft weak anomaly, a spacecraft anomaly state recognition method based on Morphological variational mode decomposition and JRD distance is proposed. First of all, the telemetry data of the spacecraft is decomposed into multi-scale modal functions with different frequencies via morphological variational modal decomposition. Then the Rényi entropy of each modal function is extracted, which is regarded as the feature of telemetry data. Finally, the recognition of spacecraft anomaly state is realized by comparing the JRD distance between the sample data and the measured data. The proposed method is verified by means of the telemetry data of the weak anomaly speed of a satellite reaction wheel. The simulation results demonstrate that the proposed method can effectively identify the anomaly of the spacecraft and has obvious advantage in recognition speed.

Method for State Recognition of Egg Embryo in Vaccines Production Based on Support Vector Machine

The method for state recognition of egg embryo in biological vaccines production based on computer image processing and support vector machine was researched. Firstly, the median filtering method was applied to eliminate noises in the images of egg embryo that were classified by the requirements of practical production, then the threshold segmentation method was used to segment the images, then the blood vessels and black blocks in the egg embryo images were selected as major characteristics to extract. The decision tree classification model with structure of binary tree was built on support vector machine, and the model was trained with the toolkit LIBSVM, the penalty factor C=2, RBF kernel parameter =3.0512e-05 and the precision of cross validation accuracy is 98.913%. Finally, 100 egg embryos were randomly selected as the test samples to do pattern recognition, and compared with manual test result.

Breaker Contact Movement State Recognition Based on Image Processing

This paper briefly describes a new method of breaker contact movement state recognition. This method is based on image target tracking. Particle filter algorithm can solve the non-linear processing problem, so it is better robustness and accuracy to the image target tracking. Using high-speed motion picture camera to obtain the circuit breaker image can solve some difficulties which are done by traditional ‘contact’ measurement methods. Experimental results show that breaker contact movement state recognition based on image processing is very effective.

COMPUTER VISION-BASED DETECTION AND STATE RECOGNITION FOR DISCONNECTING SWITCH IN SUBSTATION AUTOMATION

State recognition in disconnecting switches is important during substation automation. Here, an effective computer vision-based automatic detection and state recognition method for disconnecting switches is proposed. Taking advantage of some important prior knowledge about a disconnecting switch, the method is designed using two important features of the fixed-contact facet of such disconnecting switches. First, the Histograms of Oriented Gradients (HOG) of the fixed-contact are used to design a Linear Discriminant Analysis (LDA) target detector to position the disconnecting switches and distinguish their loci against a usual cluttered background. Then a discriminative Norm Gradient Field (NGF) feature is used to train the Support Vector Machine (SVM) state classifier to discriminate disconnecting switch states. Finally, experimental results, compared with other methods, demonstrate that the proposed method is effective and achieves a low miss rate while delivering high performance in both precision and recall rate. In addition, the adopted approach is efficient and has the potential to work in practical substation automation scenarios.

A State Recognition Approach for Complex Equipment Based on a Fuzzy Probabilistic Neural Network

Due to the traditional state recognition approaches for complex electromechanical equipment having had the disadvantages of excessive reliance on complete expert knowledge and insufficient training sets, real-time state identification system was always difficult to be established. The running efficiency cannot be guaranteed and the fault rate cannot be reduced fundamentally especially in some extreme working conditions. To solve these problems, an online state recognition method for complex equipment based on a fuzzy probabilistic neural network (FPNN) was proposed in this paper. The fuzzy rule base for complex equipment was established and a multi-level state space model was constructed. Moreover, a probabilistic neural network (PNN) was applied in state recognition, and the fuzzy functions and quantification matrix were presented. The flowchart of proposed approach was designed. Finally, a simulation example of shearer state recognition and the industrial application with an accuracy of 90.91% were provided and the proposed approach was feasible and efficient.

A sensor fusion and support vector machine based approach for recognition of complex machining conditions

During the machining process of thin-walled parts, machine tool wear and work-piece deformation always co-exist, which make the recognition of machining conditions very difficult. Existing machining condition monitoring approaches usually consider only one single condition, i.e., either tool wear or work-piece deformation. In order to close this gap, a machining condition recognition approach based on multi-sensor fusion and support vector machine (SVM) is proposed. A dynamometer sensor and an acceleration sensor are used to collect cutting force signals and vibration signals respectively. Wavelet decomposition is utilized as a signal processing method for the extraction of signal characteristics including means and variances of a certain degree of the decomposed signals. SVM is used as a condition recognition method by using the means and variances of signals as well as cutting parameters as the input vector. Information fusion theory at the feature level is adopted to assist the machining condition recognition. Experiments are designed to demonstrate and validate the feasibility of the proposed approach. A condition recognition accuracy of about 90 % has been achieved during the experiments.

State recognition method for machining process of a large spot welder based on improved genetic algorithm and hidden Markov model

Large spot welder is an important equipment in rail transit equipment manufacturing industry, but having the problem of low utilization rate and low effectlvely machining rate. State monitoring can master its operating states real time and comprehensively, and providing data support for state recognition. Hidden Markov model is a state classification method, but it is sensitive to the initial model parameters and easy to trap into a local optima. Genetic algorithm is a global searching method; however, it is quite poor at hill climbing and also has the problem of premature convergence. In this paper, proposing the improved genetic algorithm, and combining improved genetic algorithm and hidden Markov model, a new method of state recognition method named improved genetic algorithm–hidden Markov model is proposed. In the proposed method, improved genetic algorithm is used for optimizing the initial parameters, and hidden Markov model as a classifier to recognize the operating states for machining process. This method is also compared with the other two recognition methods named adaptive genetic algorithm–hidden Markov model and hidden Markov model, in which adaptive genetic algorithm is similarly used for optimizing the initial parameters, however hidden Markov model (in both methods) as a classifier. Experimental results show that the proposed method is very effective, and the improved genetic algorithm–hidden Markov model recognition method is superior to the adaptive genetic algorithm–hidden Markov model and hidden Markov model recognition method.

Multifeatures Fusion and Nonlinear Dimension Reduction for Intelligent Bearing Condition Monitoring

Condition-based maintenance is critical to reduce the costs of maintenance and improve the production efficiency. Data-driven method based on neural network (NN) is one of the most used models for mechanical components condition recognition. In this paper, we introduce a new bearing condition recognition method based on multifeatures extraction and deep neural network (DNN). First, the method calculates time domain, frequency domain, and time-frequency domain features to represent characteristic of vibration signals. Then the nonlinear dimension reduction algorithm based on deep learning is proposed to reduce the redundancy information. Finally, the top-layer classifier of deep neural network outputs the bearing condition. The proposed method is validated using experiment test-bed bearing vibration data. Meanwhile some comparative studies are performed; the results show the advantage of the proposed method in adaptive features selection and superior accuracy in bearing condition recognition.