Wavelet Packet Energy Entropy Research Articles

Gearbox degradation assessment based on a sparse representation feature and Euclidean distance technique

ABSTRACT Gearboxes have developed rapidly, and studies on gearbox degradation assessment are of great significance. In the interest of obtaining an effective gearbox degradation feature, a new methodology based on sparse representation and compressive sensing theory is proposed. The wavelet packet transform is an excellent time-frequency analysis method, and the subbands with the largest energy are selected for training the sparse dictionaries. Furthermore, the overcomplete dictionaries of each degradation state can well reflect the signal structure by calculating the compressive sensing reconstruction residuals. Based on the sparse representation feature, a Euclidean distance technique is performed to calculate the assessment rate for each degradation state. The validity and superiority of the proposed assessment method are validated by crack failure data and broken tooth failure data under three kinds of work loads. A comparison of the sparse representation feature with the features of wavelet packet energy, kurtosis, and Renyi entropy is also performed.

An Intelligent Fault Diagnosis Method of Variable Condition Gearbox Based on Improved DBN Combined With WPEE and MPE

Gear transmission is one of the most commonly used transmission methods in mechanical equipment. By analyzing the vibration data of gearbox, an improved deep belief network (DBN) algorithm for gear fault diagnosis based on wavelet packet energy entropy (WPEE) and multiscale permutation entropy (MPE) is proposed. Firstly, the vibration data of gearbox with various fault types under multiple working conditions are collected. Secondly, the energy entropy of wavelet packet and the entropy distribution of multiscale permutation are calculated respectively to form a combined feature matrix. Then, the improved threshold adaptive DBN is used to further extract the fault signal features, and finally the deep layer features are classified. By analyzing the vibration data of multi-platform gearbox, a high and stable diagnostic accuracy is obtained.

Multilayer Network from Multiple Entropies for Characterizing Gas-Liquid Nonlinear Flow Behavior

Characterizing nonlinear dynamic behaviors underlying multiphase flow has attracted considerable attention from the nonlinear research field. In this paper, the authors develop a novel multiple entropy-based multilayer network (MEMN) for exploring the complex gas-liquid two-phase flow. At first, we carry out the gas-liquid flow experiments to get the multichannel measurements. Then, MEMN is constructed based on the fusion of three nonlinear entropies, namely weighted permutation entropy (WPE), wavelet packet energy entropy (WPEE), and amplitude entropy (AE). For each derived projection network of MEMN, spectral radius and global clustering coefficient are both calculated and they allow effectively uncovering the nonlinear flow behaviors in the transition of different gas-liquid flow patterns. In addition, we perform wavelet time-frequency representation for the two typical flow patterns and the results support our findings well. All these demonstrate that our MEMN framework can effectively characterize the nonlinear evolution of gas-liquid flow from the perspective of complex network theory. And this also provides a novel idea for studying nonlinear complex systems from the observed multivariate time series.

Multi-Feature Fusion Method Based on EEG Signal and its Application in Stroke Classification.

Electroencephalogram (EEG) analysis has been widely used in the diagnosis of stroke diseases for its low cost and noninvasive characteristics. In order to classify the EEG signals of stroke patients with cerebral infarction and cerebral hemorrhage, this paper proposes a novel EEG stroke signal classification method. This method has two highlights. The first is that a multi-feature fusion method is given by combining wavelet packet energy, fuzzy entropy and hierarchical theory. The second highlight is that a suitable classification model based on ensemble classifier is constructed for perfectly classification stroke signals. Entropy is an accessible way to measure information and uncertainty of time series. Many entropy-based methods have been developed these years. By comparing with the performances of permutation entropy, sample entropy, approximate entropy in measuring the characteristic of stroke patient's EEG signals, it can be found that fuzzy entropy has best performance in characterization stroke EEG signal. By combining hierarchical theory, wavelet packet energy and fuzzy entropy, a multi-feature fusion method is proposed. The method first calculates wavelet packet energy of EEG stroke signal, then extracts hierarchical fuzzy entropy feature by combining hierarchical theory and fuzzy entropy. The experimental results show that, compared with the fuzzy entropy feature, the classification accuracy based on the fusion feature of wavelet packet energy and hierarchical fuzzy entropy is much higher than benchmark methods. It means that the proposed multi-feature fusion method based on stroke EEG signal is an efficient measure in classifying ischemic and hemorrhagic stroke. Support vector machine (SVM), decision tree and random forest are further used as the stroke signal classification models for classifying ischemic stroke and hemorrhagic stroke. Experimental results show that, based on the proposed multi-feature fusion method, the ensemble method of random forest can get the best classification performance in accuracy among three models.

Detection and Line Selection of Series Arc Fault in Multi-Load Circuit

In order to study a kind of detection and line selection method of arc fault in actual power supply and distribution lines, arc fault experiments with multi-load loop were carried out. First, five-layer decomposition of main loop current was made by using wavelet packet. The effective coefficients were selected based on the change rate of a wavelet packet energy entropy before and after the arc fault occurs. Then, an effective signal of the arc fault was reconstructed. Second, the effective signal was decomposed into seven independent modes with a variational mode decomposition method. Its time–frequency distribution was obtained by solving the Wigner–Ville distribution and performing a linear summation of each mode. Third, the time-domain and time–frequency features of arc fault were extracted by analyzing the time-domain waveform and time–frequency distribution of the effective signal. An arc fault identification and line-selection model was established by using a support vector machine optimized by particle swarm optimization and grid search. Then, the accuracy of both arc fault detection and line selection were tested. Test results indicated that the proposed method can detect effectively arc fault and select fault line accurately.

Evaluation of Rolling Bearing Performance Degradation Using Wavelet Packet Energy Entropy and RBF Neural Network

Rolling bearings are the most important parts in rotating machinery, and one of the most vulnerable parts to failure. The rolling bearing is a cyclic symmetrical structure that is stable under normal operating conditions. However, when the rolling bearing fails, its symmetry is destroyed, resulting in unstable performance and causing major accidents. If the performance of rolling bearings can be monitored and evaluated in real time, maintenance strategies can be implemented promptly. In this paper, by using wavelet packet energy entropy (WPEE), the early fault-free features of bearing and the failure samples of similar bearings are decomposed firstly, and the energy value is extracted as the original feature, simultaneously. Secondly, a radial basis function (RBF) neural network model is established by using early fault-free features and similar bearing failure characteristics. The bearing full-life data characteristics of the extracted features are added into the RBF model in an iterative manner to obtain performance degradation Indicator. Boxplot was introduced as an adaptive threshold method to determine the failure threshold. Finally, the results are verified by empirical mode decomposition and Hilbert envelope demodulation. A bearing accelerated life experiment is performed to validate the feasibility and validity of the proposed method. The experimental results show that the method can diagnose early fault points in time and evaluate the degree of bearing degradation, which is of great significance for industrial practical applications.

Adaptive Weighted VMD-WPEE Method of Power-Electronic-Circuit Multiple-Parameter-Fault Diagnosis

To improve the diagnostic accuracy rate of multiple parameter faults in widely used power electronic circuits, solving for the overlapping parameter fault characteristic signal is crucial. This paper proposes a novel approach for diagnosing the multiple parameter faults of power electronic circuits that can adaptively adjust the weights of each fault feature parameter according to the diagnostic accuracy rate to enhance their discriminability. According to the properties of multiple parameter faults in power electronic circuits, the measured signal that has been decomposed via the variational mode decomposition and wavelet packet energy entropy (VMD-WPEE) is used to extract the feature parameters. Furthermore, a classifier is selected and the threshold of the diagnostic accuracy rate is set. Then, the weight of each fault feature parameter is adjusted until the classifier yields the desired results. Finally, the proposed approach is applied to other classifiers and feature extraction methods. The high accuracy, high robustness, and advantages of the multiple-parameter-fault diagnosis method are demonstrated by the simulation and experimental results.

A Novel Rolling Bearing Fault Diagnosis and Severity Analysis Method

To improve the fault identification accuracy of rolling bearing and effectively analyze the fault severity, a novel rolling bearing fault diagnosis and severity analysis method based on the fast sample entropy, the wavelet packet energy entropy, and a multiclass relevance vector machine is proposed in this paper. A fast sample entropy calculation method based on a kd tree is adopted to improve the real-time performance of fault detection in this paper. In view of the non-linearity and non-stationarity of the vibration signals, the vibration signal of the rolling bearing is decomposed into several sub-signals containing fault information by using a wavelet packet. Then, the energy entropy values of the sub-signals decomposed by the wavelet packet are calculated to generate the feature vectors for describing different fault types and severity levels of rolling bearings. The multiclass relevance vector machine modeled by the feature vectors of different fault types and severity levels is used to realize fault type identification and a fault severity analysis of the bearings. The proposed fault diagnosis and severity analysis method is fully evaluated by experiments. The experimental results demonstrate that the fault detection method based on the sample entropy can effectively detect rolling bearing failure. The fault feature extraction method based on the wavelet packet energy entropy can effectively extract the fault features of vibration signals and a multiclass relevance vector machine can identify the fault type and severity by means of the fault features contained in these signals. Compared with some existing bearing rolling fault diagnosis methods, the proposed method is excellent for fault diagnosis and severity analysis and improves the fault identification rate reaching as high as 99.47%.

A Novel Arc Fault Detection Method Integrated Random Forest, Improved Multi-scale Permutation Entropy and Wavelet Packet Transform

Arc faults are one of the important causes of electric fires. In order to solve the problem of randomness, diversity, the concealment of series arc faults and to improve the detection accuracy, a novel arc fault detection method integrated random forest (RF), improved multi-scale permutation entropy (IMPE) and wavelet packet transform (WPT) are designed. Firstly, singular value decomposition (SVD) was applied to filter the current signal and then the high-dimensional fault features were constructed by extracting IMPE, the wavelet packet energy and the wavelet packet energy-entropy. Afterward, the high-dimensional fault features were employed to train the RF to realize the arc fault detection of different load types and the experimental results verify the effectiveness of the arc fault detection method designed in this paper. Finally, the comparative experiments demonstrates that the RF shows better performance in arc fault detection compared to the back-propagation neural network (BPNN) and least squares support vector machines (LSSVM), and that the experiments of transient events indicate that RF is able to effectively avoid incorrectly detecting different load types during the start operations and stop operations.

Fault Diagnosis of Hydraulic Seal Wear and Internal Leakage Using Wavelets and Wavelet Neural Network

The piston seal wear in hydraulic cylinder is one of the main factors that give rise to an internal leakage. This paper focuses on diagnosing piston seal wear and subsequent internal leakage from a double acting seal combination seal used in the support oil cylinder of a QY110 mobile crane. Wavelet transform is applied as a feature extractor to transform the raw oil pressure data into a feature vector consisting of wavelet packet subband energy, energy entropy, energy variance, and root mean square of the wavelet detailed coefficient $d_{4}$ . This feature vector feeds into the wavelet neural network serving as a pattern recognizer for automatically classifying the fault patterns. We demonstrate with the leakage experiment and simulation data that the proposed fault detection and identification (FDI) scheme is capable of effectively detecting and classifying the piston seal wear with excellent accuracy. Our comparison studies reveal that the proposed FDI tandem produces much more accurate result than that from back-propagation neural network. This paper is supplement to and enrichment of existing studies on fault simulation and diagnosis associated with hydraulic cylinder leakage problems.

Real-time chatter detection and automatic suppression for intelligent spindles based on wavelet packet energy entropy and local outlier factor algorithm

In this research, we proposed a real-time chatter detection and suppression module for intelligent spindle to increase machining efficiency and processing yield. For early detections of chatters, the relative wavelet packet energy entropy with high sensitivity in the high-frequency band and the local outlier factor (LOF) algorithm were utilized as chatter features and classifications, respectively. Based on the pre-obtained three-dimensional stability lobe diagram (SLD) and a LOF-based trained model, the module could real-time monitor and suppress chatter during machining processes. The module was implemented and experimentally examined with the CNC end-milling machine under five different cutting conditions for verifying the capabilities of real-time chatter identifications and suppressions. It was demonstrated that the module could detect the onset of chatter and suppress it by changing the cutting conditions to avoid damages on the surfaces of working piece due to severe chatters.

A Combined Events Recognition Scheme Using Hybrid Features in Distributed Optical Fiber Vibration Sensing System

In this paper, a high efficiency multiple events recognition scheme based on a hybrid feature extraction algorithm and a combined classifier for distributed optical fiber vibration sensing (DOFVS) system has been proposed and demonstrated. The hybrid feature vectors are extracted by using zero crossing rate, sample entropy, wavelet packet energy entropy, kurtosis, and multiscale permutation entropy. A combined classifier of support vector machine and radial basis function neural network is proposed to improve the reliability of the recognition results. The recognition result is given only when both of the two classifiers output same event types. The experimental results demonstrated that the average identification rate of five typical patterns (no intrusion, waggling the fence, climbing the fence, kicking the fence, and cutting the fence) over 97% is achieved through the combined classifier. Moreover, the whole recognition processing speed of the combined scheme is also good of real time performance, which can be limited in 1.1 s. Therefore, this kind of events recognition scheme has a quite promising application prospects in DOFVS system.

Online monitoring and diagnosis of high voltage circuit breaker faults: feature extraction analysis of vibration signals

The development of power grid system not only increases voltage and capacity, but also increases power risk. This paper briefly introduces the feature extraction method of the vibration signal of high voltage circuit breaker and support vector machine (SVM) algorithm and then analyzed the high voltage circuit breaker in three states: normal operation, fixed screw loosening and falling of opening spring, using the SVM based on the above feature extraction method. The results showed that the accuracy and precision rates of fault identification of circuit breaker were the highest by using the wavelet packet energy entropy extraction features, the false alarm rate was the lowest, and the detection time was the shortest.

Fast classification for rail defect depths using a hybrid intelligent method

In view of the fact that the traditional laser ultrasonic imaging test takes a long time and cannot achieve large area scanning of rail. This paper explores the possibility of combing the laser-ultrasonic technology and a hybrid intelligent method to fast achieve classification and evaluation of artificial rolling contact fatigue (RCF) defect in different depths. The laser ultrasonic scanning detection system is used to collect data samples from different locations of the defects quickly, and the signals are detected by an interferometer. Once the characteristic information of different rail defects is acquired and trained by Support Vector Machine (SVM), the high efficient and high-precision rail detection can be realized through the input of the feature in the detection process. The hybrid method is composed by Wavelet Packet Transform (WPT), Kernel Principal Component Analysis (KPCA) and SVM. The WPT is used to decompose the signal of surface defect in different frequency bands. The KPCA is used to eliminate the redundancy of the original feature set, thereby reducing the correlation among all the defect features. Wavelet packet time-frequency coefficient (X), energy (E) and local entropy (F) are generated and a new feature (Ynew) is created by fusing X, E and F, as a result of WPT and KPCA. Finally, a support vector machine (SVM) method is used to classify RCF defect in different depths. It implements a fast classification of small data. Compared with single features, fusion feature has the highest accuracy rate up to 98.73%.

Research on electroencephalogram emotion recognition based on the feature fusion algorithm of auto regressive model and wavelet packet entropy

Focused on the world-wide issue of improving the accuracy of emotion recognition, this paper proposes an electroencephalogram (EEG) signal feature extraction algorithm based on wavelet packet energy entropy and auto-regressive (AR) model. The auto-regressive process can be approached to EEG signal as much as possible, and provide a wealth of spectral information with few parameters. The wavelet packet entropy reflects the spectral energy distribution of the signal in each frequency band. Combination of them gives a better reflect of the energy characteristics of EEG signals. Feature extraction and fusion are implemented based on kernel principal component analysis. Six emotional states from a public multimodal database for emotion analysis using physiological signals (DEAP) are recognized. The results show that the recognition accuracy of the proposed algorithm is more than 90%, and the highest recognition accuracy is 99.33%. It indicates that this algorithm can extract the feature of EEG emotion well, and it is a kind of effective emotion feature extraction algorithm, providing support to emotion recognition.

Statements on wavelet packet energy–entropy signatures and filter influence in fault diagnosis of induction motor in non-stationary operations

Electric motors are components of great importance in mechanical systems and in the majority of the equipment used in industrial plants. The several faults that occur in the induction machines may induce severe consequences in the industrial process. Many of these faults are progressive. In this work, a contribution to the study of signal-processing techniques based on wavelet packet transform for parameter extraction of energy and entropy from vibration signals for the detection of faults in the non-stationary operation (start of the motor) is presented. Together with the wavelet transform, methods of dimensionality reduction such as principal component analysis, linear discriminant analysis, and independent components analysis are used. In addition, the use of an experimental bench shows that the model of extraction and classification proposed present high precision for fault classification.

Fault diagnosis method based on wavelet packet-energy entropy and fuzzy kernel extreme learning machine

Aiming at connatural limitations of extreme learning machine in practice, a new fault diagnosis method based on wavelet packet-energy entropy and fuzzy kernel extreme learning machine is proposed. On one hand, the presented method can extract the more efficient features using the wavelet packet-energy entropy method, and on the other hand, the sample fuzzy membership degree matrix U, weight matrix W which is used to describe the sample imbalance, and the kernel function are introduced to construct the fuzzy kernel extreme learning machine model with high accuracy and reliability. The experimental results of rolling bearing and check valve are obtained and analyzed in MATLAB 2010b. The results show that the proposed fuzzy kernel extreme learning machine method can obtain fairly or slightly better classification performance than the traditional extreme learning machine, kernel extreme learning machine, back propagation, support vector machine, and fuzzy support vector machine.

A Novel Bearing Fault Diagnosis Method Based on LMD and Wavelet Packet Energy Entropy

AbstractRolling element bearings faults may lead to fatal breakdown of machines. Therefore, it is significant to be study bearings diagnosis, and the vibration-based methods have received intensive study because vibration signals collected from bearings carry rich information on machine health conditions, and it is possible to obtain vitalcharacteristic information from the vibration signals through using signal processing techniques. This paper proposes a novel vibration-based diagnosis method about bearing faults, first, a new pattern recognition method is proposed to diagnose bearing faults through using the interval value of the spectral peak frequency in the frequency domain; second, vibration signals of different parts faults of the bearings will be processed by different algorithm for precisely extracting the fault characteristics; and third, in order to extract transient characteristics from a noisy signal, the filter need to be developed and to further improve the signal-to-noise ratio (SNR), band pass filter is designed based on the PSD of vibration signals in this paper. The vibration signals collected from rolling element bearings are used to demonstrate the performance of the proposed method, andthe results verify the effectiveness of the method in extracting fault characteristics and diagnosing faults of rolling element bearings.

Automatic detection of glottal stop in cleft palate speech

The speech therapy is essential during the whole treatment of Cleft Palate (CP) over many years. The automatic evaluation of CP speech could provide an effective assistant diagnosis. The glottal stop is a typical compensatory articulation error in CP speech. It is produced by adducting the vocal folds as a substitution during the pronunciation of oral pressure consonants. The existence of glottal stop has a great impact on speech intelligibility. In this work, an automatic glottal stop detection system is proposed. The CP speech database is collected by the Hospital of Stomatology, Sichuan University, which has the largest number of CP patients in China. This database includes extensive CP speech samples annotated by speech-language pathologists at the phoneme level. The vocabulary of this database includes all the initial consonants in Mandarin. The automatic initials and finals segmentation method is proposed firstly. Then, for the pressure initial consonants, four types of acoustic features are extracted: Mel-frequency cepstral coefficients, formants, Gammatone filtering energy feature, wavelet packet energy and Shannon entropy features. The features are processed using a k-nearest neighbor classifier. The k-fold cross validation is used to calculate the average detection accuracy, which reaches up to 93.5% using Gammatone filtering energy feature.

Study on Fault Line Selection for Single Phase Grounding in Distribution Network Based on Harmonic Wavelet Packet Energy Entropy

Study on Fault Line Selection for Single Phase Grounding in Distribution Network Based on Harmonic Wavelet Packet Energy Entropy