Showing papers on "Wavelet packet decomposition published in 2018"

Deep Residual Networks With Dynamically Weighted Wavelet Coefficients for Fault Diagnosis of Planetary Gearboxes

Abstract: One of the significant tasks in data-driven fault diagnosis methods is to configure a good feature set involving statistical parameters. However, statistical parameters are often incapable of representing the dynamic behavior of planetary gearboxes under variable operating conditions. Although the use of deep learning algorithms to find a good set of features for fault diagnosis has somewhat improved diagnostic performance, the lack of domain knowledge incorporated into deep learning algorithms has limited further improvement. Accordingly, this paper developed a variant of deep residual networks (DRNs), the so-called deep residual networks with dynamically weighted wavelet coefficients (DRN+DWWC) to improve diagnostic performance, which takes a series of sets of wavelet packet coefficients on various frequency bands as an input. Further, the fact that no general consensus has been reached as to which frequency band contains the most intrinsic information about a planetary gearbox's health status calls for “dynamic weighting layers” in the DRN+DWWC and the role of the layers is to dynamically adjust a weight applied to each set of wavelet packet coefficients to find a discriminative set of features that will be further used for planetary gearbox fault diagnosis.

NTL Detection in Electric Distribution Systems Using the Maximal Overlap Discrete Wavelet-Packet Transform and Random Undersampling Boosting

Abstract: The illegal use of electricity, defective meters, and a malfunctioning infrastructure are major causes of Non-technical losses (NTLs) in electric distribution systems Although the use of supervised machine learning techniques to detect NTLs has been widely studied, further research is needed in order to address some significant challenges (i) Given that fraudulent consumers remarkably outnumber non-fraudulent ones, the imbalanced nature of the dataset can have a major negative impact on the performance of supervised machine learning methods (ii) Given the large number of dimensions present in the time series data used for training and testing classifiers, advanced signal processing techniques are required in order to extract the most relevant information (iii) The effectiveness of classifiers must be evaluated using meaningful performance measures for imbalanced data This paper proposes a framework that addresses the three previous challenges The core of the proposed framework is the application of the maximal overlap discrete wavelet-packet transform (MODWPT) for feature extraction from time series data and the random undersampling boosting (RUSBoost) algorithm for NTL detection Moreover, our framework is evaluated using an extensive list of performance metrics Experiments show that the MODWPT combined with the RUSBoost algorithm can significantly improve the quality of NTL predictions

Online Fault Detection and Diagnosis in Photovoltaic Systems Using Wavelet Packets

Abstract: The nonlinear characteristics of a photovoltaic (PV) array, maximum power point tracking of the PV inverter, presence of blocking diodes, and lower irradiation prevent the conventional protection devices to trip under certain faults and lead to reduced system efficiency and fire hazards. Moreover, the PV characteristics under certain partial shading conditions are similar to that under fault conditions. Hence, it is imperative to detect faults and differentiate faults from the partial shading condition to avoid false tripping of the system. This paper proposes a simple fault detection method using the available data of array voltage and current by means of wavelet packets. The proposed scheme is simulated using MATLAB/Simulink and is experimentally tested on a 1.6-kW $\text{4}\,\times \,\text{4}$ PV array to validate its performance.

Fault Diagnosis Based on Weighted Extreme Learning Machine With Wavelet Packet Decomposition and KPCA

Abstract: Fault diagnosis has received considerable attention because its implementation can effectively prevent costly and even catastrophic downtime. However, quickly identifying faults and accurately obtaining diagnosis results from a feature set of rotating machinery are still a problem. To this end, this paper proposes an effective method based on a weighted extreme learning machine (WELM) with wavelet packet decomposition (WPD) and kernel principal component analysis (KPCA). The feature set affecting classification accuracy can be obtained using WPD and KPCA. By taking feature reliability into consideration, a new type of improvement to the extreme learning machine (ELM), i.e., WELM, is proposed by associating the hidden layer and input layer with a weight matrix. The WELM model can help in guaranteeing a quick and an accurate identification of fault status. To verify the superiority of the fault identification speed and accuracy of the proposed method, results from other methods, namely, using the sensitive features based on WPD and KPCA with ELM, a back-propagation neural network, and a support vector machine, were compared. The experimental results indicate that the proposed method can effectively improve the accuracy and quickly diagnose the fault. The average accuracy of fault classification could reach 95.45%, and the computation time of WELM was only 0.0156 s.

Application of stationary wavelet entropy in pathological brain detection

Abstract: Labeling brain images as healthy or pathological cases is an important procedure for medical diagnosis. Therefore, we proposed a novel image feature, stationary wavelet entropy (SWE), to extract brain image features. Meanwhile, we replaced the feature extraction procedure in state-of-the-art approaches with the proposed SWE. We found the classification performance improved after replacing wavelet entropy (WE), wavelet energy (WN), and discrete wavelet transform (DWT) with the proposed SWE. This proposed SWE is superior to WE, WN, and DWT.

Fault feature extraction of rolling element bearings based on wavelet packet transform and sparse representation theory

Abstract: Extracting reliable features from vibration signals is a key problem in machinery fault recognition. This study proposes a novel sparse wavelet reconstruction residual (SWRR) feature for rolling element bearing diagnosis based on wavelet packet transform (WPT) and sparse representation theory. WPT has obtained huge success in machine fault diagnosis, which demonstrates its potential for extracting discriminative features. Sparse representation is an increasingly popular algorithm in signal processing and can find concise, high-level representations of signals that well matches the structure of analyzed data by using a learned dictionary. If sparse coding is conducted with a discriminative dictionary for different type signals, the pattern laying in each class will drive the generation of a unique residual. Inspired by this, sparse representation is introduced to help the feature extraction from WPT-based results in a novel manner: (1) learn a dictionary for each fault-related WPT subband; (2) solve the coefficients of each subband for different classes using the learned dictionaries and (3) calculate the reconstruction residual to form the SWRR feature. The effectiveness and advantages of the SWRR feature are confirmed by the practical fault pattern recognition of two bearing cases.

A Wavelet-Based Statistical Approach for Monitoring and Diagnosis of Compound Faults With Application to Rolling Bearings

Abstract: This paper proposes a wavelet-based statistical signal detection approach for monitoring and diagnosis of bearing compound faults at an early stage. The bearing vibration signal is decomposed by an orthonormal discrete wavelet transform to obtain its energy dispersions at multiple levels. We investigate the statistical properties of the decomposed signal energy under both the normal and faulty conditions, based on which a generalized likelihood ratio test is developed. An exponentially weighted moving average control chart is then constructed to detect faults at an early stage. Simulation studies and a real case study are conducted to demonstrate the effectiveness of the proposed method. Furthermore, the comparison studies show that the proposed method outperforms the empirical mode decomposition method and Hilbert envelope spectrum analysis method. Note to Practitioners —This paper is motivated by the problem of monitoring and diagnosis of compound faults in rolling bearings at the early stage, which are seldom considered in existing methods. In this paper, we propose a new approach by using statistical signal detection method and wavelet transform to handle the fault signals. This work aims at monitoring vibration signals and diagnosing fault types. Our simulation results show the proposed approach outperforms existing methods, especially at an early stage. Our future work will aim at improving the method’s sensitivity in distinguishing faults similar to each other.

Online chatter detection of the end milling based on wavelet packet transform and support vector machine recursive feature elimination

Abstract: Chatter is a common state in the end milling, which has important influence on machining quality. Early chatter detection is a prerequisite for taking effective measures to avoid chatter. However, there are still many difficulties in the feature extraction of chatter detection. In this article, a novel online chatter detection method in end milling process is proposed based on wavelet packet transform (WPT) and support vector machine recursive feature elimination (SVM-RFE). The measured vibration signal in the machining process was preprocessed by WPT. The original feature set of chatter composed of ten time-domain and four frequency-domain feature parameters was obtained via calculating the reconstructed signal. Then feature weights are computed by SVM-RFE, and the obtained feature ranking list was to indicate their different importance in chatter. The optimal feature subset was selected according to the prediction accuracy. The proposed method is described and applied to incipient chatter over conventional methods in identifying the transition from a stable to unstable state. Some milling tests were conducted and the experiment results was shown that the impulse factor and onestep autocorrelation function were the sensitive chatter features.

Empirical Wavelet Transform Based Features for Classification of Parkinson’s Disease Severity

Abstract: Parkinson's disease (PD) is a type of progressive neurodegenerative disorder that has affected a large part of the population till now. Several symptoms of PD include tremor, rigidity, slowness of movements and vocal impairments. In order to develop an effective diagnostic system, a number of algorithms were proposed mainly to distinguish healthy individuals from the ones with PD. However, most of the previous works were conducted based on a binary classification, with the early PD stage and the advanced ones being treated equally. Therefore, in this work, we propose a multiclass classification with three classes of PD severity level (mild, moderate, severe) and healthy control. The focus is to detect and classify PD using signals from wearable motion and audio sensors based on both empirical wavelet transform (EWT) and empirical wavelet packet transform (EWPT) respectively. The EWT/EWPT was applied to decompose both speech and motion data signals up to five levels. Next, several features are extracted after obtaining the instantaneous amplitudes and frequencies from the coefficients of the decomposed signals by applying the Hilbert transform. The performance of the algorithm was analysed using three classifiers - K-nearest neighbour (KNN), probabilistic neural network (PNN) and extreme learning machine (ELM). Experimental results demonstrated that our proposed approach had the ability to differentiate PD from non-PD subjects, including their severity level - with classification accuracies of more than 90% using EWT/EWPT-ELM based on signals from motion and audio sensors respectively. Additionally, classification accuracy of more than 95% was achieved when EWT/EWPT-ELM is applied to signals from integration of both signal's information.

A Multisensor Fusion Method for Tool Condition Monitoring in Milling

Abstract: Tool fault diagnosis in numerical control (NC) machines plays a significant role in ensuring manufacturing quality. Tool condition monitoring (TCM) based on multisensors can provide more information related to tool condition, but it can also increase the risk that effective information is overwhelmed by redundant information. Thus, the method of obtaining the most effective feature information from multisensor signals is currently a hot topic. However, most of the current feature selection methods take into account the correlation between the feature parameters and the tool state and do not analyze the influence of feature parameters on prediction accuracy. In this paper, a multisensor global feature extraction method for TCM in the milling process is researched. Several statistical parameters in the time, frequency, and time–frequency (Wavelet packet transform) domains of multiple sensors are selected as an alternative parameter set. The monitoring model is executed by a Kernel-based extreme learning Machine (KELM), and a modified genetic algorithm (GA) is applied in order to search the optimal parameter combinations in a two-objective optimization model to achieve the highest prediction precision. The experimental results show that the proposed method outperforms the Pearson’s correlation coefficient (PCC) based, minimal redundancy and maximal relevance (mRMR) based, and Principal component analysis (PCA)-based feature selection methods.

A Robust Multibit Multiplicative Watermark Decoder Using a Vector-Based Hidden Markov Model in Wavelet Domain

Abstract: The vector-based hidden Markov model (HMM) is a powerful statistical model for characterizing the distribution of the wavelet coefficients, since it is capable of capturing the subband marginal distribution as well as the inter-scale and cross-orientation dependencies of the wavelet coefficients. In this paper we propose a scheme for designing a blind multibit watermark decoder incorporating the vector-based HMM in wavelet domain. The decoder is designed based on the maximum likelihood criterion. A closed-form expression is derived for the bit error rate and validated experimentally with Monte Carlo simulations. The performance of the proposed watermark detector is evaluated using a set of standard test images and shown to outperform the decoders designed based on the Cauchy or generalized Gaussian distributions without or with attacks. It is also shown that the proposed decoder is more robust against various kinds of attacks compared with the state-of-the-art methods.

Feature extraction of power transformer vibration signals based on empirical wavelet transform and multiscale entropy

Abstract: To achieve an effective feature extraction for power transformer vibration signals, the authors propose a method for signal feature extraction based on empirical wavelet transform (EWT) and multiscale entropy (MSE). First, transformer vibration signals are decomposed into several empirical wavelet functions (EWFs) with the method of EWT. Then, the frequency characteristics of signals are demonstrated in the time-frequency representation by applying a Hilbert transform to each EWF component. Finally, in order to quantify the extracted features, the MSEs of components being highly correlated with the original signals are calculated to construct the eigenvectors of transformer vibration signals. Several experiments are presented showing the effectiveness of this method compared with the classic empirical mode decomposition method.

A Novel Fault Diagnostic Approach for DC-DC Converters Based on CSA-DBN

Abstract: Effective fault diagnosis for mission-critical and safety-critical systems has been an essential and mandatory technique to reduce failure rate and prevent unscheduled shutdown. In this paper, to realize fault diagnosis for a closed-loop single-ended primary inductance converter, a novel optimization deep belief network (DBN) is presented. First, wavelet packet decomposition is adopted to extract the energy values from the voltage signals of four circuit nodes, as the fault feature vectors. Then, a four-layer DBN architecture including input and output layers is developed. Meanwhile, the number of neurons in the two hidden layers is selected by the crow search algorithm (CSA) with training samples. Not only the hard faults such as open-circuit faults and short-circuit faults but also the soft faults such as the component degradation of power MOSFET, inductor, diode, and capacitor are considered in this study. Finally, these fault modes are isolated by CSA-DBN. Compared with the back-propagation neural network and support vector machine fault diagnosis methods, both simulation and experimental results show that the proposed method has a higher classification accuracy that proves its effectiveness and superiority to the other methods.

Fault diagnosis method of rolling bearing using principal component analysis and support vector machine

Abstract: To effectively extract the fault feature information of rolling bearings and improve the performance of fault diagnosis, a fault diagnosis method based on principal component analysis and support vector machine was presented, and the rolling bearings signals with different fault states were collected. To address the limitation on effectively dealing with the raw vibration signals by the traditional signal processing technology based on Fourier transform, wavelet packet decomposition was employed to extract the features of bearing faults such as outer ring flaking, inner ring flaking, roller flaking and normal condition. Compared with the previous literature on fault diagnosis using principal component analysis (PCA) and support vector machine (SVM), one-to-one and one-to-many algorithms were taken into account. Additionally, the effect of four kernel functions, such as liner kernel function, polynomial kernel function, radial basis function and hyperbolic tangent kernel function, on the performance of SVM classifier was investigated, and the optimal hype-parameters of SVM classifier model were determined by genetic algorithm optimization. PCA was employed for dimension reduction, so as to reduce the computational complexity. The principal components that reached more than 95 % cumulative contribution rate were extracted by PCA and were input into SVM and BP neural network classifiers for identification. Results show that the fault feature dimensionality of the rolling bearing is reduced from 8-dimensions to 5-dimensions, which can still characterize the bearing status effectively, and the computational complexity is reduced as well. Compared with the raw feature set, PCA has a higher fault diagnosis accuracy (more than 97 %), and a shorter diagnosis time relatively. To better verify the superiority of the proposed method, SVM classification results were compared with the results of BP neural network. It is concluded that SVM classifier achieved a better performance than BP neural network classifier in terms of the classification accuracy and time-cost.