Detail Coefficients Research Articles

Remote Fault Identification and Analysis in Electrical Distribution Network Using Artificial Intelligence

This research describes a method for wavelet decomposition and machine learning-based fault site classification in a radial power distribution network. The first statistical observation is produced using wavelet decomposition and wavelet-based detailed coefficients in terms of Kurtosis and Skewness parameters. For this objective, six distinct machine learning methods are deployed. They are evaluated and compared using unknown data sets with varying degrees of unpredictability. One approach has been shown to be the most accurate in locating the location of the problem bus.

SEMG Signals Identification Using DT And LR Classifier by Wavelet-Based Features

In the recent era of technology, biomedical signals have been attracted lots of attention regarding the development of rehabilitation robotic technology. The surface electromyography (SEMG) signals are the fabulous signals utilized in the field of robotics. In this context, SEMG signals have been acquired by twenty-five right-hand dominated healthy human subjects to discriminate the various hand gestures. The placement of SEMG electrodes has been done according to the predefined acupressure point of required hand movements. After the SEMG signal acquisition, pre-processing and noise rejection have been performed. The de-noising and four levels of SEMG signal decomposition have been accomplished by discrete wavelet transform (DWT). In this article, the third and fourth-level detail coefficients have been utilized for time-scale feature extractions. The performance of ten time-scale features has been evaluated and compared to each other with the three-fold cross-validation technique by using a Decision Tree (DT) and Linear Regression (LR) classifier. The results demonstrated that the DT classifier classification accuracy was found superior to the LR classifier. By using the DT classifier technique 96.3% accuracy has been achieved, with all combined features as a feature vector.

Removal of strong noise in magnetotelluric data using grey wolf optimized wavelet threshold

AbstractTraditional magnetotelluric signal processing usually uses time–frequency transformation method. Wavelet is also a time–frequency transformation method that used to suppress the magnetotelluric noise. However, the selection of the threshold is very significant, and the unsuitable threshold will lead to excessive distortion of the reconstructed signal. Thus, we propose a method for magnetotelluric noise suppression using grey wolf optimized wavelet threshold. First, the magnetotelluric signal is decomposed by wavelet with appropriate wavelet basis and decomposition layers. The generalized cross‐validation criterion is used as the fitness function of grey wolf optimizer algorithm, which optimizes the threshold of each decomposition layer. Then, the detail coefficients of each layer and the maximum layer of the approximation coefficients are used in the optimized threshold. Next, the inverse wavelet transform is performed. Finally, the noise contour is obtained through iteratively searching for the optimal threshold, and the useful magnetotelluric signal is reconstructed. Simulation experiments and measured magnetotelluric data processing show that the large‐scale interference can effectively be suppressed, and the reconstructed magnetotelluric signal retains the more abundant low‐frequency of useful information. Compared with the remote reference method, fixed threshold method and Birge–Massart layered threshold method, the proposed method realizes the wavelet denoising with adaptive threshold selection in the magnetotelluric noise suppression. The results obtained show smoother and more continuous apparent resistivity–phase curves, which verifies the effectiveness of the optimization.

Condition-Based Health Monitoring of Electrical Machines Using DWT and LDA Classifier

In the industry, continuous health monitoring of electric motors is considered as an essential requirement. The continuous operation of the electric motor may cause malfunctions and addressing them timely is a critical challenge. The development of an efficient health monitoring system based on the identification of electrical motor faults is on great demand. This paper addresses the fault detection technique using discrete wavelet transform (DWT) algorithm for continuous health monitoring of electric motor-based systems. The faults have been detected through Motor Current Signature Analysis (MCSA) series procedures using the proposed method. Concurrently, the wavelet transform algorithm produces frequency-based spectrum related to the stator current parameters to accomplish the fault classification. This study provides an analysis of three motor faults of Phase imbalance, Rotor misalignment, and High contact resistance (HCR). DWT has the ability to categorize the input signals into approximate coefficient state for low frequency signals and detailed coefficient state for high frequency signals. In this research, this technique is used to detect faults because it is able of processing signals of very low frequency, and effectively deal with intermittent sharp signals that appear frequently during processing. DWT technique based on conditional monitoring of an induction motor with precise detailed coefficients and more skilled at light loads given on a motor-shaft with relatively fast execution time compared to FFT. Furthermore, the comparison of healthy and faulty induction motors has been compiled by Linear Discriminant Analysis (LDA) technique, a sub-application of MATLAB, and used for faults management purposes. LDA in comparison with PCA gives more perfect results. In this research, different faults have been detected with 100% accuracy using LDA classifier. The implementation of the proposed scheme will be beneficial in avoiding faults by ensuring that preemptive measures are taken timely against these faults, and the production of industries is protected from revenue losses.

PSMFF: A progressive series-parallel modality feature filtering framework for infrared and visible image fusion

Current researches in the infrared and visible image fusion task tend to focus on designing decomposition or fusion schemes for improving the feature preservation performance. However, raw images captured by multi-sensors from extreme conditions may suffer from noise, low contrast and loss of textural and structural details. These influences are difficult to be eliminated by most of the current fusion schemes. In this paper, a robust progressive series-parallel modality feature filtering framework (PSMFF) is proposed to provide a promising solution for 24/7 infrared and visible image integration. After fully considering modality-specific characteristics of infrared and visible images, the first-stage enhancement module is constructed as a dual solution system that combines the pixel-level attention-aware module and fidelity-driven intensity enhancement to overcome the noise and low contrast of source images. Moreover, to further excavate and integrate the latent inherent information of each modality, the second-stage enhancement and fusion Scheme based on the gradient-domain guide filter is proposed to realize a better extraction and aggregation of significant features from the decomposed base and detail coefficients. Notably, the proposed PSMFF method can fulfill the low-light fusion task and preserves more informative edge and texture details to obtain a satisfactory fusion result. Extensive experiments demonstrate that the proposed algorithm is superior to the state-of-the-art methods in the performance of subjective and objective evaluation.

Robust R-peak detection in an electrocardiogram with stationary wavelet transformation and separable convolution

R-peak detection is an essential step in analyzing electrocardiograms (ECGs). Previous deep learning models reported their performance primarily in a single database, and some models did not perform at the highest levels when applied to a database different from the testing database. To achieve high performances in cross-database validations, we developed a novel deep learning model for R-peak detection using stationary wavelet transform (SWT) and separable convolution. Three databases (i.e., the MIT-BIH Arrhythmia [MIT-BIH], the Institute of Cardiological Technics [INCART], and the QT) were used in both the training and testing models, and the MIT-BIH ST Change (MIT-BIH-ST), European ST-T, TELE and MIT-BIH Noise Stress Test (MIT-BIH-NST) databases were further used for testing. The detail coefficient of level 4 decomposition by SWT and the first derivative from filtered ECGs were used for model inputs, and the interval of 150 ms centered at marked peaks was used for labels. Separable convolution with atrous spatial pyramidal pooling was selected as the model’s architecture, and noise-augmented waveforms of 5.69 s duration (2048 size in 360 Hz) were used in training. The model performance was evaluated using cross-database validation. The F1 scores of the peak detection model were 0.9994, 0.9985, and 0.9999 in the MIT-BIH, INCART, and QT databases, respectively. When the above three databases were pooled, the F1 scores were 0.9993 for fivefold cross-validation and 0.9991 for cross-database validation. The model performance remained high for MIT-BIH-ST, European ST-T, and TELE, with F1 scores of 0.9995, 0.9988, and 0.9790, respectively. The model performance when trained by severe noise augmentation increased for the MIT-BIH-NST database (F1 scores from 0.9504 to 0.9759) and decreased for the MIT-BIH database (F1 scores from 0.9994 to 0.9991). The present SWT and separable convolution-based model for R-peak detection yields a high performance even for cross-database validations.

Monitoring effects of heavy metal stress on biochemical and spectral parameters of cotton using hyperspectral reflectance.

Monitoring heavy metal pollution in agricultural ecosystems is crucial to ensure environmental safety. Heavy metals interfere with plants' biochemical characteristics, such as chlorophyll content and photosynthesis, andalso influence leaves' spectral properties. Spectral changes caused by heavy metal stress can easily be measured using proximal sensing or in-field spectroscopy. This research utilizes a combined approach of biochemical and spectral characteristics to evaluate cottoncrops' performance under different heavy metal (Pb & Cd)stress after artificial contamination with the metal under study. A detailed study of spectroscopy and lab-based measurements for chlorophyll and heavy metal content during the crop's growth cycle revealed some significant findings. Results indicated that the chlorophyll pigments decreased significantly with increased heavy metal levels. Pb accumulation is high in cotton as compared to Cd. The most sensitive stage for the accumulation of Pb is the initial vegetative stageof cotton. The transfer factor from soil to plant was higher for Pb, indicating the feasibility of growing cotton in Pb-contaminated soil. The spectral measurement showed no characteristic changes in standard reflectance spectra due to heavy metal stress. Wavelet decomposition of reflectance spectra amplified the changes indicating Pb stress in cotton during the initial vegetative stage. The significant correlation of greater than - 0.70 between the reconstructed detail wavelet coefficients at the third level of the decomposition in the wavelength range of 651-742nm suggested that Pb stress caused spectral changes in near-infrared and visible ranges in cotton plants. The effects of Cd stress on the cotton plant were negligible due to less absorption. Thus, detailed wavelet coefficients at the third level of decomposition in the range of 651-742nm are a potential indicator of Pb stress. The results of this study can provide a basis for quantifying heavy metal stress in a particular region.

A New Technique Based on AMID Using Adaptive Thresholding for Ultrasound Speckle Reduction

A new technique based on adaptive multi-resolution image decomposition (AMID) using adaptive thresholding for ultrasound speckle reduction has been proposed in this paper. The noisy image is first filtered to reduce large multiplicative noise. The image is then log-transformed before decomposing it to its approximation and detailed coefficients using AMID algorithm. The approximation coefficients are appropriately filtered while adaptive thresholding is applied to the detail coefficients. The image is reconstructed back from all the modified coefficients to obtain the denoised image. This method combining adaptive decomposition with adaptive thresholding makes it adaptive to both signal and noise components, thereby retaining the edge details in the denoised image effectively. The proposed technique performance is tested on synthetic as well as real ultrasound images with the noise of different variances. The experimental results show that the proposed algorithm gives better performance than the other state-of-the-art methods in terms of edge keeping index (EKI), correlation coefficient (CC), figure of merit (FOM), peak signal-to-noise ratio (PSNR), structural similarity (SSIM) and signal-to-noise ratio (SNR) for synthetic images. The algorithm gives better performance in terms of equivalent number of looks (ENL) and mean-to-variance ratio (MVR) for real ultrasound images.

Deep malware detection framework for IoT-based smart agriculture

The advancement in smart agriculture through the Internet of Things (IoT) devices has increased the risk of cyber-attacks. Most of the existing malware detection techniques are unable to detect new variants of malware and suffer from poor accuracy. To overcome the challenges of new malware, this research work proposes a novel three-phase Deep Malware Detection (DMD) framework based on the fusion of Discrete Wavelet Transform (DWT) and Generative Adversarial Network (GAN) named as DMD-DWT-GAN, for IoT-based Smart Agriculture (IoT-SA). This work applied DWT for multiresolution analysis by decomposing the image into Approximation coefficients (Ac) and Detail coefficients (Dc). Finally, a lightweight Convolutional Neural Network (CNN) is employed for in-depth analysis of the malware family. The performance of the proposed framework is evaluated using two benchmark datasets such as IoT malware and Malimg. The proposed framework has achieved 99.99% accuracy on both datasets and is better than the state-of-art models.

Intelligent fuzzy system for automatic artifact detection and removal from EEG signals

The EEG signals were used in many medical and technological applications such as diagnosis of diseases, rehabilitation of disabled peoples, preventive healthcare, BCI (brain computer interface) systems. EEG signal is prone to the physiological and non-physiological artifacts which severely affect them and lead to its misinterpretation. An automatic method and/or algorithm; for handling EEG artifacts; is proposed. The proposed method is based on three statistical parameters (entropy, kurtosis and skewness), fuzzy inference system (FIS) and stationary wavelet transform (SWT). Each incoming EEG epoch is described using these three statistical parameters. Based on the extracted statistical parameters, the designed FIS decides if an epoch is artifactual or not. Then SWT is used to decompose the EEG epoch into detail and approximation coefficients. To reduce the effect of artifact removal, we propose to use other fuzzy inference systems, which allow to select the contaminated wavelet coefficients. The universal thresholding method is then applied to the corrupted coefficients. Finally, the inverse SWT applies to the thresholded and non-corrupted coefficients to restore the cleaned EEG signal. The performance of the proposed method in terms of amount of artifact removal and signal distortion is evaluated in three scenarios: fully simulated, semi-simulated, and real artifactual EEG data. The comparison of our method with some existing state-of-the-art methods shows the superiority of our method over others in terms of performance and computational time.

The crack detection of acoustic metamaterials using a weighted mode shape-wavelet-based strategy

With the extensive application of acoustic metamaterial (AM) in the mechanical industry, the crack detection of AM becomes increasingly crucial. In this paper, a weighted mode shape-wavelet-based strategy coupled with edge-based smoothed finite element (ES-FEM) is presented for damage detection of AM. Firstly, the multi-order mode shapes of cracked AM are computed by ES-FEM, and the obtained mode shape data is further interpolated and extended. Then, the multi-order detail coefficients of smoothed mode shapes are calculated through the discrete wavelet transform (DWT). Finally, the damage index and crack width estimator are established from the weighted multi-order detail coefficients to locate the crack and predict crack width, respectively. Numerical results demonstrate that the traditional modal-wavelet based crack detection strategy is very ambiguous while the proposed ES-FEM coupled weighted crack detection strategy is very effective for crack location and crack size prediction of AM.

Errors Incurred in Local Convective Heat Transfer Coefficients Obtained through Transient One-Dimensional Semi-Infinite Conduction Modeling: A Computational Heat Transfer Study

In typical turbulent flow problems, detailed heat transfer coefficient (h) maps obtained through short-duration experiments are based on inverse heat transfer methods that take the wall temperatures measured via liquid crystals or infrared thermography as input, and an error minimization routine is adopted to determine the best value of h that satisfies the wall temperature temporal evolution under a certain change in fluid temperature. A common practice involves modeling the solid as a one-dimensional semi-infinite medium by selecting the solid material that has low thermal conductivity and low thermal diffusivity. However, in certain flow scenarios, the neglection of the lateral heat diffusion may lead to significant errors in the deduced h values. It is imperative to understand the reasons behind large errors that may be incurred by using the 1D heat conduction assumption in order to accurately determine high-resolution h maps for better heat exchanger designs in a wide range of thermal management applications. This paper presents a computational heat transfer study on different jet impingement scenarios to demonstrate the errors incurred in the determination of h when calculated under the assumption of one-dimensional (1-d) heat conduction into a solid. To this end, three different cases are studied: (a) single jet, (b) array jet (theoretical distribution), (c) array jet (experimental distribution), along with three different mainstream temperature evolution profiles representing step change, moderately fast transient and slow transient nature of flow driving the heat transfer in the solid. A known distribution of heat transfer coefficient (“true h”) for each of the three cases is considered, and three-dimensional transient heat diffusion equations were solved to populate temperatures of each node in the solid at every time step. It is found that stagnation zones’ h1d calculations were lower than the “true h” while the low heat transfer zones exhibited significantly higher h1d compared to the “true h”. For the array jet (experimental distribution) case, it was observed that errors can be as high as 10% in certain low heat transfer zones. Different data reduction procedures, configurations, and conditions explored in this study indicate that a suitable balance can be achieved if shorter time durations in transient experiments are used as a reference for tracking in h1d calculations to keep the deviations from the “true h” low.

Sleep stage classification in EEG signals using the clustering approach based probability distribution features coupled with classification algorithms

Sleep scoring is one of the primary tasks for the classification of sleep stages in Electroencephalogram (EEG) signals. Manual visual scoring of sleep stages is time-consuming as well as being dependent on the experience of a highly qualified sleep expert. This paper aims to address these issues by developing a new method to automatically classify sleep stages in EEG signals. In this research, a robust method has been presented based on the clustering approach, coupled with probability distribution features, to identify six sleep stages with the use of EEG signals. Using this method, each 30-second EEG signal is firstly segmented into small epochs and then each epoch is divided into 60 sub-segments. Each sub-segment is decomposed into five levels by using a discrete wavelet transform (DWT) to obtain the approximation and detailed coefficient. The wavelet coefficient of each level is clustered using the k-means algorithm. Subsequently, features are extracted based on the probability distribution for each wavelet coefficient. The extracted features then are forwarded to the least squares support vector machine classifier (LS-SVM) to identify sleep stages. Comparisons with several existing methods are also made in this study. The proposed method for the classification of the sleep stages achieves an average accuracy rate of 97.4%. It can be an effective tool for sleep stages classification and can be useful for doctors and neurologists for diagnosing sleep disorders.

Single-scale convolution wavelet feature optimization classification model based on electrocardiogram coded image

Premature beats are a cardiovascular disease, which can lead to complications of other diseases. An electrocardiograph (ECG) is the main tool for detecting premature beats, but accurate detection of premature beats still faces challenges. This work designs a single-scale convolution wavelet feature optimization (SSCWFO) classification model that is based on a single-scale convolutional neural network (SSCNN) model and optimized by Linear Discriminant Analysis (LDA) algorithm. First, the wavelet coefficient features are extracted from 12-lead ECG signals using Daubechies5, and then the 5th level detail coefficient features are transformed into images using Gramian Angular Difference Fields (GADFs). Thereafter the single-scale features are extracted by the SSCNN model. Next, the LDA algorithm was used to maximum inter-class distance and the minimum intra-class distance of each category. Finally, the Gaussian naive Bayes is used to classify-three types of signals. The results after LDA optimization show that the accuracy, precision, sensitivity, F1-score, and area under the ROC curve (AUC) of the SSCWFO model are improved to 91.12%, 92.29%, 92.55%, 92.13%, and 0.9197, respectively. Also, the network architectures of Resnet34, Resnet18, and LeNet 5 are used in this work for a comparison; their accuracy is 75.93%, 76.88%, and 68.48%, respectively. This shows that the method can effectively distinguish normal signals and two types of premature beat diseases and is helpful to the effective classification of diseases of premature beats.

Continuous wavelet estimation for multivariate fractional Brownian motion

In this paper, we propose a method using continuous wavelets to study the multivariate fractional Brownian motion through the deviations of the transformed random process to find an efficient estimate of Hurst exponent using eigenvalue regression of the covariance matrix. The results of simulations experiments shown that the performance of the proposed estimator was efficient in bias but the variance get increase as signal change from short to long memory the MASE increase relatively. The estimation process was made by calculating the eigenvalues for the variance-covariance matrix of Meyer’s continuous wavelet details coefficients.

A new Fuzzy&Wavelet-based adaptive thresholding method for detecting PQDs in a hydrogen and solar-energy powered EV charging station

This study presents a hybrid fuzzy decision-maker (FDM) and un-decimated wavelet transform (UWT)-based method for detecting power quality disturbances (PQDs) in a developed hydrogen and solar energy-powered electric vehicle (EV) charge station. The proposed adaptive FDM&UWT-based hybrid method eliminated the lack of performance of threshold-based signal analysis methods in noise-containing signals and it is implemented for a reliable PQD detection and integration in a developed microgrid. Also, the proposed method has eliminated the need for a processing-intensive filtering process to reduce noise from the signal. With this adaptive approach, detection errors in boundary conditions in threshold value methods are avoided and at the same time, cost and computational burden are minimized by using only the peak values in the detail coefficients of the voltage signal. The mean test accuracy is 96.13% for the FDM method using pyramidal UWT in noise-free conditions. Besides, the pyramidal UWT-FDM has a mean classification accuracy of 94.96% under 20–40 dB high-level noise conditions. The effectiveness of the UWT-FDM method is also tested using an experimental setup. The mean test accuracy for experimental data is 96.66%.

Diagnosis of induction motor failures using discrete wavelet transform of an auxiliary winding voltage

Over last years, great attention has been developed to avoid the machines breakdown especially in the squirrel cage induction motor which suffer from different failures. The aim of this paper is to present a new method for the diagnosis of rotor bar breakage and end ring in induction machine fed by an inverter based on discrete wavelet transform applied on the voltage of an auxiliary winding as a new monitoring signature. The expression of the auxiliary winding voltage related to a small coil inserted between two stator phases is presented. The study is focused on the high-level signals of approximation and details coefficients. Thus, the evolution of any frequency of interest in the waveform is given in this paper. The method is validated by simulations of four broken bars and end ring cases under unloaded and loaded machine.

A new signal processing approach/method for classification of power quality disturbances

The disadvantage of conventional signal processing methods used for Power Quality (PQ) classification is that their mathematical calculations are complex. Therefore, it is necessary to develop simple, effective, and fast signal processing tools to analyze the PQ signal. In this study, a new signal processing method with simple mathematical operations based on the Integer Factor (IF) down sampling/Approximation Derivatives (AD) was developed to analyze PQ signals. IF was employed to analyze the signals at various sampling frequencies, while the AD method was employed to obtain different degrees of detail and approximation coefficients for the signals. In order to prove that the IF/AD signal processing method can perform a fast and detailed signal processing analysis, the IF/AD method and the Discrete Wavelet Transforms (DWT)/Multiresolution method, which is one of the basic signal processing methods, were compared. It was determined that the developed signal processing method has analyzed PQ disturbances in a shorter time and in more detail than DWT/Multiresolution method. After it was determined that the IF/AD method was effective in the analysis of PQ signals, a new classification method based on the IF-AD signal processing approach, the Slime Mould feature selection algorithm, and Support Vector Machine was developed. The classification method was applied to synthetic signals with noise, real-time data generated in the laboratory and PQ online data. The results proved that the developed classification method was successful. Therefore, this study will provide a different approach to PQ classification methods and non-stationary signal analysis studies.

Wavelet sparsity enhancement for extracting transient vibration signatures of bearing structural damages

Wavelet methods are widely used in mechanical transient vibration signature detection and fault diagnosis. Undesirable artifacts (e.g., spurious noise spikes and pseudo-Gibbs components) are serious issues for the mainstream methods, probably resulting in inaccurate analysis results. For this reason, a new wavelet sparsity enhancement methodology is proposed to achieve artifact-free extraction of bearing transient vibration signatures. The wavelet sparsity is enhanced through two aspects—that is, wavelet basis design and wavelet coefficient processing. Specifically, we use a new family of wavelets called fractional B-spline wavelets on bearing fault signal analysis and propose the adaptive undecimated fractional spline wavelet transform which addresses the shift-invariant problem and also permits to customize an optimal wavelet basis according to the signal itself adaptively. Meanwhile, we introduce a two-step wavelet processing method including a nonlinear operator and a generalized hard-thresholding (with symmetric or asymmetric thresholds determined automatically by wavelet coefficients at each level). Moreover, unlike the most existing wavelet methods, the approximation coefficients are also processed along with the detailed coefficients to remove the possible low-frequency noise. The final transient vibration signatures are reconstructed with the processed coefficients and would be sparser, more accurate, and almost free of artifacts. The validity of the methodology is verified with the analysis results of vibration signals measured from fault-injection experiment and industrial wind turbine transmission system. The comparisons highlight the advantages of the methodology over several common methods in suppressing artifacts and extracting the sparse transient vibration signatures of bearing structural damages.

Detection of Alzheimer's Dementia by Using Signal Decomposition and Machine Learning Methods.

Dementia is one of the most common neurological disorders causing defection of cognitive functions, and seriously affects the quality of life. In this study, various methods have been proposed for the detection and follow-up of Alzheimer's dementia (AD) with advanced signal processing methods by using electroencephalography (EEG) signals. Signal decomposition-based approaches such as empirical mode decomposition (EMD), ensemble EMD (EEMD), and discrete wavelet transform (DWT) are presented to classify EEG segments of control subjects (CSs) and AD patients. Intrinsic mode functions (IMFs) are obtained from the signals using the EMD and EEMD methods, and the IMFs showing the most significant differences between the two groups are selected by applying previously suggested selection procedures. Five-time-domain and 5-spectral-domain features are calculated using selected IMFs, and five detail and approximation coefficients of DWT. Signal decomposition processes are conducted for both 1 min and 5 s EEG segment durations. For the 1 min segment duration, all the proposed approaches yield prominent classification performances. While the highest classification accuracies are obtained using EMD (91.8%) and EEMD (94.1%) approaches from the temporal/right brain cluster, the highest classification accuracy for the DWT (95.2%) approach is obtained from the temporal/left brain cluster for 1 min segment duration.