Transient Feature Extraction Research Articles

Transient Signal Analysis Using Parallel Time-Frequency Manifold Filtering for Bearing Health Diagnosis

It is crucial to identify and extract the weak transient features embedded in the vibration signals for bearing health monitoring and fault diagnosis. However, due to the macro-structural disturbance and background noise interference, it is not easy to mine the transient features, especially at the apparent failure stage. Meanwhile, the actual mechanism of bearing fault detect can be not simply expressed by the formulated theory models without consideration of the actual physical collision process. To overcome these issues, motivated by the merits of time-frequency manifold (TFM), this paper proposes a new transient feature extraction method, called parallel time-frequency manifold (PTFM) filtering, by simultaneously using TFM-based reconstruction with TFM-based filter in parallel for transient feature extraction. First, to improve the computational efficiency of TFM, two-dimensional discrete wavelet transform is employed on the raw time-frequency distribution (TFD) with image compression. TFM learning is later used to mine the principle manifolds from those approximation sub-images. Then, the amplitudes of the raw time-frequency image can be reconstructed by TFM feature bases while the desired location of time-frequency feature can be captured by TFM morphology filter in a process of image morphology. With raw time-frequency phases in a series of inverse processes, the de-noised signal can be finally synthesized from these filtered images. The proposed method accomplishes a natural manifold feature denoising by combining the sparse theory with image morphology, and demonstrates attractive prospects in the following three aspects: signal de-noising with a self-learning mode in the view of image morphology processing combined with sparse theory, fault diagnosis with in-band noise/close interference removal, and machine health monitoring with capability in capturing sensitive failure information. Simulations and experiments confirmed the effectiveness of the proposed PTFM filtering method in noise suppression and feature enhancement, which is valuable for bearing health monitoring and diagnosis applications.

SSVEP Transient Feature Extraction and Rapid Recognition Method Based on Bistable Stochastic Resonance.

Steady-state Visual Evoked Potential, SSVEP), as the most commonly used communication paradigm for non-implantable Brain-Computer Interface (BCI), boasts the advantages of unnecessity of training, noise immunity and periodicity. The traditional SSVEP extraction methods can effectively identify the target frequency contained in original EEG, however, the required data length usually lasts a few seconds. In this paper, bistable stochastic resonance (BSR) is applied to SSVEP extraction. BSR is very sensitive to amplitude mutation and frequency fluctuation of the input signal, making the output difference can be used for the detection of the target frequency. The processing results illustrate that the proposed method not only has a high recognition accuracy, but also effectively shortens the recognition time, thus improving the calculating speed. Therefore, SSVEP extraction based on BSR has a higher information transmission rate (ITR), which is more suitable for the real-time BCI system.

A New Family of Model-Based Impulsive Wavelets and Their Sparse Representation for Rolling Bearing Fault Diagnosis

The localized faults of rolling bearings can be diagnosed by the extraction of the impulsive feature. However, the approximately periodic impulses may be submerged in strong interferences generated by other components and the background noise. To address this issue, this paper explores a new impulsive feature extraction method based on the sparse representation. According to the vibration model of an impulse generated by the bearing fault, a novel impulsive wavelet is constructed, which satisfies the admissibility condition. As a result, this family of model-based impulsive wavelets can form a Parseval frame. With the model-based impulsive wavelet basis and Fourier basis, a convex optimization problem is formulated to extract the repetitive impulses. Based on the splitting idea, an iterative thresholding shrinkage algorithm is proposed to solve this problem, and it has a fast convergence rate. Via the simulated signal and real vibration signals with bearing fault information, the performance of the proposed approach for repetitive impulsive feature extraction is validated and compared with the noted spectral kurtosis method, the optimized spectral kurtosis method based on simulated annealing, and the resonance-based signal decomposition method. The results demonstrate its advantage and superiority in weak repetitive transient feature extraction.

Transients Extraction Based on Averaged Random Orthogonal Matching Pursuit Algorithm for Machinery Fault Diagnosis

Sparse representation based on matching pursuit (MP) algorithm is one of the effective methods for the extraction of weak feature contaminated by heavy noise. However, the optimal iterative threshold and iterations of MP algorithm are difficult to be determined and the sparsest representation is difficult to obtain for pursuit algorithms. In order to reduce the influence of the threshold and iterations on the performance of MP and obtain appropriate results without seeking the sparsest representation, a new transient feature extraction technique named averaged random orthogonal MP (AROMP) algorithm is proposed. In the proposed method, random orthogonal MP algorithm, which is a greedy algorithm to match the atoms with probability, is utilized repeatedly to generate a group of competitive representations for the mechanical vibration signal. Then by averaging these solutions, the estimated representation vector can be obtained to represent the vibration signal and then transients can be extracted from the noisy signal. The simulation study and experimental analysis show that transients can be extracted effectively from the noisy vibration signal. And comparison results between the proposed method and orthogonal MP show that the proposed algorithm is less dependent on iterations and can obtain a better performance for transients extraction. Comparisons between the proposed algorithm and spectral kurtosis also show the superiority of the proposed method.

A Transient Fault Recognition Method for an AC-DC Hybrid Transmission System Based on MMC Information Fusion

At present, the research is still in the primary stage in the process of fault disturbance energy transfer in the multilevel modular converter based high voltage direct current (HVDC-MMC). An urgent problem is how to extract and analyze the fault features hidden in MMC electrical information in further studies on the HVDC system. Aiming at the above, this article analyzes the influence of AC transient disturbance on electrical signals of MMC. At the same time, it is found that the energy distribution of electrical signals in MMC is different for different arms in the same frequency bands after the discrete wavelet packet transformation (DWPT). Renyi wavelet packet energy entropy (RWPEE) and Renyi wavelet packet time entropy (RWPTE) are proposed and applied to AC transient fault feature extraction from electrical signals in MMC. Using the feature extraction results of Renyi wavelet packet entropy (RWPE), a novel recognition method is put forward to recognize AC transient faults using the information fusion technology. Theoretical analysis and experimental results show that the proposed method is available to recognize transient AC faults.

Cosmos supersonic reentry analysis: A case study using traditional and mobile infrasound arrays

On August 31, 2015, the fiery reentry of the Soviet Cosmos 1315 was sighted over the Hawaiian Islands. The supersonic reentry of the satellite was detected with both traditional infrasound arrays located on the Big Island, Maui and Kauai, and with iPods using the RedVox infrasound recorder on the Big Island and Oahu. We apply array-processing techniques to reconstruct the trajectory of the satellite. In addition, we have developed a standardized multi-resolution framework that uses the Gabor limit to define a scaled set of frequencies and windows for transient feature extraction. Metrics are evaluated over a set of standardized log-scaled time windows in order to characterize waveforms. I describe a method to detect transients in infrasonic arrays with low signal to noise where beam forming is not possible, and apply it to the analysis of signals produced by the supersonic Cosmos 1315 reentry.

Weak transient fault feature extraction based on an optimized Morlet wavelet and kurtosis

Aimed at solving the key problem in weak transient detection, the present study proposes a new transient feature extraction approach using the optimized Morlet wavelet transform, kurtosis index and soft-thresholding. Firstly, a fast optimization algorithm based on the Shannon entropy is developed to obtain the optimized Morlet wavelet parameter. Compared to the existing Morlet wavelet parameter optimization algorithm, this algorithm has lower computation complexity. After performing the optimized Morlet wavelet transform on the analyzed signal, the kurtosis index is used to select the characteristic scales and obtain the corresponding wavelet coefficients. From the time-frequency distribution of the periodic impulsive signal, it is found that the transient signal can be reconstructed by the wavelet coefficients at several characteristic scales, rather than the wavelet coefficients at just one characteristic scale, so as to improve the accuracy of transient detection. Due to the noise influence on the characteristic wavelet coefficients, the adaptive soft-thresholding method is applied to denoise these coefficients. With the denoised wavelet coefficients, the transient signal can be reconstructed. The proposed method was applied to the analysis of two simulated signals, and the diagnosis of a rolling bearing fault and a gearbox fault. The superiority of the method over the fast kurtogram method was verified by the results of simulation analysis and real experiments. It is concluded that the proposed method is extremely suitable for extracting the periodic impulsive feature from strong background noise.

Sparse representation based on local time–frequency template matching for bearing transient fault feature extraction

The transients caused by the localized fault are important measurement information for bearing fault diagnosis. Thus it is crucial to extract the transients from the bearing vibration or acoustic signals that are always corrupted by a large amount of background noise. In this paper, an iterative transient feature extraction approach is proposed based on time–frequency (TF) domain sparse representation. The approach is realized by presenting a new method, called local TF template matching. In this method, the TF atoms are constructed based on the TF distribution (TFD) of the Morlet wavelet bases and local TF templates are formulated from the TF atoms for the matching process. The instantaneous frequency (IF) ridge calculated from the TFD of an analyzed signal provides the frequency parameter values for the TF atoms as well as an effective template matching path on the TF plane. In each iteration, local TF templates are employed to do correlation with the TFD of the analyzed signal along the IF ridge tube for identifying the optimum parameters of transient wavelet model. With this iterative procedure, transients can be extracted in the TF domain from measured signals one by one. The final signal can be synthesized by combining the extracted TF atoms and the phase of the raw signal. The local TF template matching builds an effective TF matching-based sparse representation approach with the merit of satisfying the native pulse waveform structure of transients. The effectiveness of the proposed method is verified by practical defective bearing signals. Comparison results also show that the proposed method is superior to traditional methods in transient feature extraction.

Low frequency (LF) RFID sensors and selective transient feature extraction for corrosion characterisation

In this paper, Low Frequency (LF) RFID sensing system is used to characterise marine atmospheric corrosion on steel samples. Considering the conductivity and permeability variation in the corrosion layer, optimisation of sensing sensitivity and communication distance of commercial RFID tags placed on both coated and uncoated samples (exposure duration between 1 and 12 months) for impedance matching and appropriate operation frequencies are analysed. To balance the communication distance between RFID reader and tag and to match the impedance variation of RFID tags on metallic samples, a concept of wireless power transfer (WPT) through strongly coupled magnetic resonance (SCMR) method at RFID is applied. Static and transient features are extracted and selected to characterise corrosion. To enhance the sensing sensitivity to corrosion characterisation, further experimental studies have been carried out using ferrite sheets. The proposed LF RFID sensing system has been validated using commercial RFID tags with different feature extractions and selection of features for corrosion characterisation.

Condition assessment for automatic tool changer based on sparsity-enabled signal decomposition method

Automatic tool changer (ATC) is one of the key basic parts in CNC machining centers, and the globoidal indexing cam and the groove cam are the functional units for tool changing. Thus the condition monitoring is important for highly efficient and precision machining. In this paper, a condition monitoring system is constructed for the ATC, especially for the globoidal indexing cam, including vibration signal acquisition, fault feature extraction and localization, and condition assessment. In the constructed system, sparsity-enabled signal decomposition method is introduced to extract transient component and reduce noises in the complex vibration signals, and the transient component is always a key feature for fault localization. Simulation study shows that the sparsity-enabled signal decomposition method is effective in transient feature extraction. The experimental application in condition assessment for the ATC demonstrates that the constructed condition monitoring system has the potential to assess the working condition of the ATC in practical application.

Sparse representation of transients in wavelet basis and its application in gearbox fault feature extraction

Vibration signals from a defective gearbox are often associated with important measurement information useful for gearbox fault diagnosis. The extraction of transient features from the vibration signals has always been a key issue for detecting the localized fault. In this paper, a new transient feature extraction technique is proposed for gearbox fault diagnosis based on sparse representation in wavelet basis. With the proposed method, both the impulse time and the period of transients can be effectively identified, and thus the transient features can be extracted. The effectiveness of the proposed method is verified by the simulated signals as well as the practical gearbox vibration signals. Comparison study shows that the proposed method outperforms empirical mode decomposition (EMD) in transient feature extraction.

Development of an Improved Time–Frequency Analysis-Based Nonintrusive Load Monitor for Load Demand Identification

In a smart house connected to a smart grid via advanced metering infrastructure, a nonintrusive load monitor (NILM) that identifies individual appliances by disaggregating composite electric load signal from the minimal number of sensors installed at the main distribution board in the field can be regarded as a part of a home/building energy management system. This type of load monitoring technique, not only for domestic but also for industrial applications, is relevant to electricity energy management and conservation issues. In this paper, an improved time-frequency analysis-based NILM composed of three system components, including data acquisition, transient feature extraction, and load identification, is proposed. The improved NILM proposed in this paper incorporates a multiresolution S-transform-based transient feature extraction scheme with a modified 0-1 multidimensional knapsack algorithm-based load identification method to identify individual household appliances that may either be energized simultaneously or be identified under similar real power consumption. For the load identification process, an ant colony optimization algorithm is employed to perform combinatorial search that is formulated as a modified 0-1 multidimensional knapsack problem. As shown from the experimental results, the improved NILM strategy proposed in this paper is confirmed to be feasible.

Sparse Representation De-Noising Based on Morlet Wavelet Basis and its Application for Transient Feature Extraction

Signals with multiple transients are often encountered with much noise in engineering. The transient feature extraction has always been the key issue for signal analysis. A new signal de-noising method combining sparse representation and Morlet wavelet basis is proposed for signal de-noising and feature extraction. Simulation study concerning multiple transients signal shows the effectiveness of this method in transient feature extraction. The efficiency of this de-noising method is also verified by its application to extract fault signature for gearbox.

Transient Signal Feature Extraction Based on Box Dimension and the Largest Lyapunov Dimension

Cognitive radio is one kind of wireless communication technology, which is to effectively improve the spectrum efficiency by the way of wireless spectrum sensing. Besides of varieties of security issues of the traditional wireless communication, it also has some new problems, such as Primary User Emulation (PUE) attack. This essay presents that based on the fractal dimension of the transient signal characteristics extraction to protect the PUE attack from the cognitive wireless networks, and then extract the fractal dimension of wireless transmitter through experiments to provide reliable characteristic parameters for the next signal recognition, finally to prevent PUEs attack.

Research on Simulation System for Power Transient Signal Analysis Based on Wavelet Theory

For the purpose of power system transient signal analysis, recognition and classification respectively ,this paper focuses on the study of electric power transient signal acquisition, signal preprocessing method and transient signal feature extraction .Based on the wavelet theory and some simulation experiments, the process of transient signal analysis is studied. And it makes a primary development of electric power transient signal analysis automatic simulation system by the simulation software. In this research, the wavelet denoising method is used and to extract the transient signal feature, which can be used in detecting transient occurrence time. Finally, the system for transient signal analysis is developed and has the wide application fields.

Research on Simulation System for Power Transient Signal Analysis Based on Wavelet Theory

For the purpose of power system transient signal analysis, recognition and classification respectively ,this paper focuses on the study of electric power transient signal acquisition, signal preprocessing method and transient signal feature extraction .Based on the wavelet theory and some simulation experiments, the process of transient signal analysis is studied. And it makes a primary development of electric power transient signal analysis automatic simulation system by the simulation software. In this research, the wavelet denoising method is used and to extract the transient signal feature, which can be used in detecting transient occurrence time. Finally, the system for transient signal analysis is developed and has the wide application fields.

Modern development of an Adaptive Non-Intrusive Appliance Load Monitoring system in electricity energy conservation

Traditionally, electricity energy consumption of a residence can be monitored through the watt-hour meter installed at the panel. However, the power consumption of individual appliances cannot be identified directly. To track and control the energy consumption of each appliance, many meters are needed. This approach results in high financial costs. To reduce the installation costs, the Non-Intrusive Appliance Load Monitoring (NIALM) approach was proposed recently. In this paper, a novel Adaptive Non-Intrusive Appliance Load Monitoring (ANIALM) system that integrates appliance energizing and de-energizing transient feature extraction methods with soft-computing techniques is developed to keep track of the energy consumption of each appliance. The energizing and de-energizing transient responses of appliances can be captured through the analyses of ANIALM. Two recognizers, k-Nearest Neighbor Rule (k-NNR) and Back-Propagation Artificial Neural Network (BP-ANN), are used to identify different types of appliances and their operation status under different single-load and multiple-load operation scenarios. The Artificial Immune Algorithm (AIA) with the Fisher criterion is employed to adaptively adjust the feature parameters in order to improve the identification performance of recognizers when a new type of appliance is added for monitoring. From the experimental results obtained in different actual environments, the proposed ANIALM system is confirmed that it is able to identify the operation status of appliances. Also, although the generalization of both recognizers is similar and excellent, the k-NNR recognizer used by the ANIALM system is preferred from the aspects of recognizers’ identification mechanism and training performance due to its simplicity in computation and implementation.

Automaton Fault Diagnosis Based on Motion Morphology and Information Entropy

Online monitoring and fault diagnosis is an important link of guaranteeing the equipment smooth operation and reliable working, which receives general concern. This subject uses the strategy of combining theoretical research and experimental research, and establishes a set of automaton fault diagnosis theory and method based on motion morphology and information entropy. It solves the following problems, the weak fault signals in the short-time and transient vibration response signals are easy to be drowned in practical application, effective and sensitive characteristic parameters are difficult to be extracted, accurate positioning of fault and real-time diagnosis are difficult to be realized. Use the motion morphology and information entropy to put forward new ideas and methods for the short-time and transient vibration signals analysis processing and feature extraction, and is applied in artillery automaton field, which expands the research scope of mechanical fault diagnosis subject.

Quantitative evaluation of angular defects by pulsed eddy current thermography

Pulsed eddy current (PEC) thermography employs a combination of PEC and thermographic non-destructive testing (NDT) techniques. This study considers the capabilities of PEC thermography for obtaining quantitative information about cracks set at an angle to the surface. The investigation is implemented by simulating the transient thermal distribution for angular slots, via time-stepping 3D finite element analysis (FEA), with the experimental work undertaken for verification. A slope inclination feature of the transient temperature distribution has been extracted and presented for estimating the angle of slots that is independent of slot depth and length inside the sample. With the identification of the slot angle, quantification of the length/depth of the slot inside the sample can then be made through a maximum temperature amplitude feature. Experimental studies have been undertaken for evaluation of the numerical simulation and transient feature extraction methods.

Sensor Selection for Machine Olfaction Based on Transient Feature Extraction

Machine olfaction devices, which are often called electronic noses (e-noses), are gaining favor for odor assessment applications in several industrial sectors, such as beverage, perfumery, and food. From a design point of view, the number of sensors in these devices for a particular odor application should be minimized without degrading classification accuracy. This paper deals with selecting sensors for e-noses to make small portable devices with fast response times and reduced cost possible. Prior research efforts have been reported in the open literature and have shown that many advantages can be gained by properly selecting the input features before forwarding to a pattern classification algorithm. This selection process can reduce the dimensionality of the feature space, remove redundant and irrelevant features, speed up classification, and improve classification performance. In this paper, the transient features of an array of sensors obtained by applying a multiresolutional approximation technique from the discrete wavelet transform (DWT) are investigated to search for an optimal sensor array to be implemented in the e-nose system. A genetic algorithm is adapted to tailor a gas sensor array for two different odor data sets (coffee and soda). From the experimental results, the input features obtained by applying the DWT to the transient sensor responses not only provide a significant reduction in the number of sensors when compared to traditional features but also improve the classification rate to near 100%.