Synchrosqueezing Wavelet Transform Research Articles

Synchrosqueezing Voices Through Deep Neural Networks for Horizon Interpretation

Horizon picking stands as a crucial element in reservoir characterisation, yet it remains a labour-intensive process. The manual interpretation of horizons across thousands of vertical seismic slices in a 3D seismic survey significantly further amplifies the time and effort invested in this task. While several automatic methods have been developed for extracting horizons in seismic images, their effectiveness can be compromised in the presence of interruptions in lateral continuity, such as faults and noise. Additionally, closely spaced horizons pose a challenge, making it even more difficult to accurately depict their exact locations. For tracking the horizon surfaces through a 3D seismic volume, it is necessary to exploit other seismic attributes extracted from the 3D seismic data. We proposed to use spectral voice components together with the original seismic amplitudes to track target horizon surfaces. We generated the time-frequency spectrum using a high-resolution method namely the synchrosqueezing wavelet transform (SWT) method, and the real part of the complex SWT spectrum is the voice component. We imported the spectral voice component and the seismic amplitude into a neural network. A framework of deep convolutional neural network (dCNN) was adopted for tracking horizon surfaces within a 3D seismic volume. We demonstrated this application on a field seismic dataset where closely spaced thin layers are within a complex faulted formation with noisy and low signal to noise ratio seismic data. The integration of amplitude and phase within the voice component attribute demonstrates its efficacy in enhancing the quality of the generated horizons, particularly when compared to using only seismic amplitude for this task. A field data example from the F3 dataset showcases the capability of our method in accurately delineating horizons across fault surfaces and in close proximity to unconformities. This surpasses the current limitations of existing horizon-picking methods.

Bearing Fault Diagnosis under Varying Work Conditions Based on Synchrosqueezing Transform, Random Projection, and Convolutional Neural Networks

Bearings are critical components in rotating machinery, and their failure can lead to costly repairs and downtime. To prevent such failures, it is important to detect and diagnose bearing faults early. In recent years, deep-learning techniques have shown promise for detecting and diagnosing bearing faults automatically. While these algorithms can all achieve diagnostic accuracy of over 90%, their generalizability and robustness in complex, extreme variable loading conditions have not been thoroughly validated. In this paper, a feature extraction method based on Synchro-squeezing Wavelet Transform (SSWT), Random projection (RP), and deep learning (DL) is presented. To fulfill the data requirements of neural networks, data augmentation is initially utilized to augment the size of the original data. Subsequently, the SSWT technique is employed to convert the signals from the Time domain to the Time-Frequency domain, resulting in the conversion of the 1-D signal to a 2-D feature image. To decrease the complexity of deep learning computation, data preprocessing involves utilizing Random projection to reduce feature dimensionality. The final step involves constructing a Convolutional Neural Network (CNN) model that can identify fault features from the obtained Time-Frequency images and perform accurate fault classification. By utilizing the CWRU and IMS datasets to evaluate the method, the study demonstrates that the suggested approach outperforms advanced techniques in terms of both diagnostic accuracy and robustness.

Forced Oscillation Source Location of Bulk Power Systems Using Synchrosqueezing Wavelet Transform

Forced Oscillation Source Location of Bulk Power Systems Using Synchrosqueezing Wavelet Transform

A subject-independent portable emotion recognition system using synchrosqueezing wavelet transform maps of EEG signals and ResNet-18

ObjectiveDesigning a portable Brain-Computer Interface (aBCI) using EEG signals is challenging due to the numerous channels, though not all are vital for emotional recognition. We aimed to simplify this by creating a two-channel portable aBCI using advanced time-frequency analysis and deep learning. MethodOur approach involved utilizing the time-frequency analysis named synchrosqueezing wavelet transform (SSWT), which provides better frequency resolution for fluctuations of EEG signal than common wavelet transform. Using the ResNet-18 Convolutional Neural Network, we fine-tuned for sadness and happiness classification. The two best channels were identified across four databases: SEED-IV, SEED-V, SEED-GER, and SEED-FRA, using the Leave-One-Subject-Out method. ResultsFinally, we achieved an average accuracy over sad and happy emotions using the SSWT-ResNet18 model of 76.66%, 78.12%, 81.25%, and 75.00% for the SEED-IV, SEED-V, SEED-GER, and SEED-FRA databases, respectively. ConclusionOverall, our study demonstrates the potential for developing a rapid aBCI by utilizing a precise time–frequency method and deep learning technique from the least number of channels. SignificanceOur approach has promising implications for future real-world applications in emotional recognition.

A Case Study of Synchrosqueezing Wavelet Transform to Conventional Logging

In this paper, the synchronous squeezing wavelet transform (SSWT) is used to process the conventional logging curve, and the spectral characteristics of logging signal are obtained. The results show that the spectral characteristics of SSWT logging in tight sandstone reservoir are more obvious than that in depth domain, and have different characteristics for different reservoir fluid properties, which is helpful to identify the fluid properties of complex reservoirs.

Adaptive synchrosqueezing wavelet transform for real-time applications

Time-Frequency Analysis is an important tool that is widely used in engineering, applied sciences, and industrial applications. Several techniques have emerged in the last decade to increase the time-frequency resolution of such representations. However, in many cases, these techniques imply a heavy computational burden, which is particularly troublesome for embedded devices, edge computing, or online applications in general. In this paper, we introduce an adaptive algorithm based on the Synchrosqueezing Wavelet Transform, aimed at implementations on systems with constrained computational power. We test the performance of the algorithm on both synthetic and real-world signals. Additionally, we present a software architecture to implement the data processing algorithm keeping in mind the portability of the solution to different devices. A companion Python package has been made available for public use.

Oscillatory time–frequency concentration for adaptive bearing fault diagnosis under nonstationary time-varying speed

Being a promising post-processing solution, application of time–frequency concentration (TFC) suffers in an adaptive fault diagnosis scenario due to non-adaptive extraction of bearing fault signature from noise-associated vibration signals under nonstationary time-varying speed. Aiming at solving this problem, a novel method called oscillatory time–frequency concentration (OTFC) is proposed by oscillation based adaptive extraction of time-varying bearing fault signature with the help of tunable Q factor wavelet transform (TQWT). OTFC can effectively suppress unwanted noise while adaptively reveals the time-varying fault characteristic frequency (FCF) and corresponding harmonics with excellent readability. Two distinct numerical simulations and experimental case studies have been utilized to verify the performance of the proposed OTFC. The results show that the proposed OTFC not only performs better than the conventional TFC method namely synchrosqueezing wavelet transform (SST) but also shows superior performance than traditional continuous wavelet transform (CWT) and advanced TFC method namely multitaper synchrosqueezing wavelet transform (MST).

SEMG signal filtering study using synchrosqueezing wavelet transform with differential evolution optimized threshold

Most gesture recognition studies based on surface electromyography (sEMG) signals focus on filtering, in which the lack of diversity for considered noises can still be the problem. In this work, a denoising method based on Synchrosqueezing Wavelet Transform with Differential Evolution optimized threshold (DEOT-SWT) is proposed. The sEMG signals of ten gestures with three mixed noises, including power line interference (PLI), baseline drift (BW), and white Gaussian noise (WGN), are firstly investigated and filtered by DEOT-SWT, which are collected from seven subjects by utilizing two wearable sEMG signal sensors. Then, the most commonly used Hudgins time-domain feature set is extracted for recognizing ten gestures. Three metrics are adopted to evaluate filtering performance: signal-to-noise ratio (SNR), root mean square error and R-squared value. The gesture recognition accuracy is utilized to verify the practical effect of DEOT-SWT in sEMG-based gesture recognition applications. The results of the experiments demonstrate that the DEOT-SWT algorithm accomplishes desirable denoising performance with an average recognition accuracy of 95.95% (±3.88) in comparison to the classic Infinite Impulse Response (IIR) algorithm and the empirical mode decomposition (EMD) algorithm.

TS-LSSVM: Triple sparse least squares support vector machine for residual oxygen concentration detection of encapsulated pharmaceutical vials

Accurate measurement of residual oxygen concentration in encapsulated pharmaceutical vials is adequate to ensure the quality of inner sterile preparations. However, the critical characteristic signal is feeble and covered by enormous environmental interference in the actual production. Inspired by structural sparse learning, we propose a novel prediction model in this paper, triple sparse least squares support vector machine (TS-LSSVM), in which the production priors are deeply excavated, and the feature, sample, and structure sparsity are realized simultaneously by redefining the objective function. In addition, the selection of support vectors can be adjusted adaptively according to the time-varying environmental noise, so as to ensure the reliability of long-term operation. First, the time–frequency components containing the prior knowledge are extracted based on the synchro squeezing wavelet transform (SSWT). Then, a triple sparse learning strategy is designed, which can accurately eliminate redundant wavelet coefficients and adaptively select training samples. Finally, a strict and fast optimization solution is proposed under the alternating direction method of multipliers (ADMM) framework. Experimental results on public and practical datasets prove the superiority of TS-LSSVM.

A Novel Fault Diagnosis Method Based on SWT and VGG-LSTM Model for Hydraulic Axial Piston Pump

Since the hydraulic axial piston pump is the engine that drives hydraulic transmission systems, it is widely utilized in aerospace, marine equipment, civil engineering, and mechanical engineering. Operating safely and dependably is crucial, and failure poses a major risk. Hydraulic axial piston pump malfunctions are characterized by internal concealment, challenging self-adaptive feature extraction, and blatant timing of fault signals. By completely integrating the time-frequency feature conversion capability of synchrosqueezing wavelet transform (SWT), the feature extraction capability of VGG11, as well as the feature memory capability of the long short-term memory (LSTM) model, a novel intelligent fault identification method is proposed in this paper. First, the status data are transformed into two dimensions in terms of time and frequency by using SWT. Second, the depth features of the time–frequency map are obtained and dimensionality reduction is carried out by using the deep feature mining capability of VGG11. Third, LSTM is added to provide the damage identification model for long-term memory capabilities. The Softmax layer is utilized for the intelligent evaluation of various damage patterns and health state. The proposed method is utilized to identify and diagnose five typical states, including normal state, swash plate wear, sliding slipper wear, loose slipper, and center spring failure, based on the externally observed vibration signals of a hydraulic axial piston pump. The results indicate that the average test accuracy for five typical state signals reaches 99.43%, the standard deviation is 0.0011, and the average test duration is 2.675 s. The integrated model exhibits improved all-around performance when compared to LSTM, LeNet-5, AlexNet, VGG11, and other typical models. The proposed method is validated to be efficient and accurate for the intelligent identification of common defects of hydraulic axial piston pumps.

Hierarchical extreme puzzle learning machine-based emotion recognition using multimodal physiological signals

Detection of exact emotions through multi-modal physiological signals provides relevant information for different processes. Numerous computational approaches have been presented for the precise analysis of emotion types. But due to some problems like ruined signal quality, increased time consumption, and the necessity of high storage space, classification accuracy’s efficiency worsens. Hence, this research classified multi-modal physiological signals based on machine and deep learning (DL) models. The proposed work implements the Hierarchical Extreme Puzzle Learning Machine (HEPLM) approach to classify the actual output of embedded emotions. The proposed work comprises four steps: pre-processing, signal-to-image conversion, feature extraction, and classification. Pre-processing is carried out using Savitzky-Golay smoothing filtering (SGF) for the removal of noise and to increase signal quality. Hybrid wavelet scattering with Synchro squeezing Wavelet Transform approach converts the signal into an image. In feature extraction process, the valuable features are extracted using ResNet-152 and the Inception v3 model, whereas the features are combined through an ensemble approach. HEPLM is used in the final classification process, combining Puzzle Optimization Algorithm (POA) and Hierarchical Extreme Learning Machine (HELM) to reduce feature dimensionality and improve classification accuracy. The dataset adopted in the proposed work is Wearable Stress and Affect Detection (WESAD) to collect multi-modal physiological signals. The presentation of the projected work is assessed with metrics like accuracy, recall, precision, F1 score, kappa, and so on. The proposed effort demonstrates better results of emotion classification when compared to the existing methods by holding 96.29% of accuracy.

Tensor Denoising Assisted Time-Reassigned Synchrosqueezing Wavelet Transform for Gear Fault Diagnosis

The gear fault signal is characterized by strong and/or weak impulsive features and is often accompanied by strong background noise, which makes the gear fault feature extraction more difficult. Hence, this paper proposes a new time-frequency analysis method called the tensor denoising (TD) assisted time-reassigned synchrosqueezing wavelet transform (TD-TSWT). First, to improve the accuracy of subsequent time-frequency analysis, a TD method based on high-order singular value decomposition is proposed as a preprocessing technique according to the cyclostationary characteristics of gear faults and the Hankel matrix. Meanwhile, combined with the advantages of sample entropy calculation, the concept of using sample entropy to determine the reasonable singular order is proposed. Second, to propose TSWT for revealing gear fault features. TSWT by introducing continuous wavelet transform to obtain the basis function which is more similar to gear fault impulses. Moreover, according to the properties of the gear fault signal, the group delay estimation operator in the frequency domain is designed for TSWT, which is more suitable for the gear fault feature extraction. Finally, the results of simulations and engineering cases show that the method provides a useful tool for the diagnosis of gear faults.

A Novel Hybrid Method to Predict PM2.5 Concentration Based on the SWT-QPSO-LSTM Hybrid Model

PM2.5 concentration is an important indicator to measure air quality. Its value is affected by meteorological factors and air pollutants, so it has the characteristics of nonlinearity, irregularity, and uncertainty. To accurately predict PM2.5 concentration, this paper proposes a hybrid prediction system based on the Synchrosqueezing Wavelet Transform (SWT) method, Quantum Particle Swarm Optimization (QPSO) algorithm, and Long Short-Term Memory (LSTM) model. First, the original data are denoised by the SWT method and taken as the input of the prediction model. Then, the main parameters of the LSTM model are optimized by global search based on the QPSO algorithm, which solves the problems of slow convergence and local extremum of traditional parameter training algorithms. Finally, the PM2.5 daily concentration data of Chengdu, Shijiazhuang, Shenyang, and Wuhan are predicted by the proposed SWT-QPSO-LSTM model, and the prediction results are compared with those of single prediction models and hybrid prediction models. The experimental results show that the proposed model achieves higher prediction precision and lower prediction error than other models.

A common-reflection-point gather random noise attenuation method based on the synchrosqueezing wavelet transform

Common-reflection-point (CRP) gather is one extensively used prestack seismic data type. However, CRP suffers more noise than a poststack seismic data set. The events in the CRP gather are always flat, and the effective signals from neighboring traces in the CRP gather have similar forms not only in the time domain, but also in the time-frequency (TF) domain. Therefore, we first use the synchrosqueezing wavelet transform (SSWT) to decompose seismic traces to the TF domain because the SSWT has better TF resolution and reconstruction properties. Then, we use the similarity of neighboring traces to smooth and threshold the SSWT coefficients in the TF domain. Finally, we used the modified SSWT coefficients to reconstruct the denoised traces for the CRP gather. Synthetic and field data examples indicate that our method can effectively attenuate random noise with a better attenuation performance than the commonly used principal component analysis, [Formula: see text] filter, and the continuous wavelet transform method.

Spatial-Domain Synchrosqueezing Wavelet Transform and Its Application to Seismic Ground Roll Suppression

Spatial-Domain Synchrosqueezing Wavelet Transform and Its Application to Seismic Ground Roll Suppression

Random noise attenuation in seismic data using Hankel sparse low-rank approximation

The Hankel matrix’s low-rank property derived from the noise-free seismic data describing a few linear events and has been successively leveraged in many low-rank seismic data de-noising approaches. In such rank reduction methods, the typical scheme is to determine the best low-rank estimation of the formulated Hankel matrix, and then obtain the de-noised data. However, if the noisy data has been rearranged for the low-rank approximation in a Hankel matrix, it is usually not precisely low-rank. In the presented research, we propose a multivariate generalization of the minimax-concave penalty (MCP) function inducing sparsity on seismic data in the time-space domain. Initially obtained sparse representation of data would be decomposed into semi low-rank and the sparse components with the best approximate of noisy measurement matrix would be defined. This would be performed through the low-rank matrix extraction by optimal (re)weighting of the singular vectors of the observed matrix. The efficiency of the proposed method was evaluated on synthetic and real land data examples. Results were also compared with the state-of-art methods such as the non-local means (NLM), the Optimum Shrinkage Sparse Low-Rank estimation, the Optimum Shrinkage Synchrosqueezing Wavelet Transform and the damped rank reduction (DRR) methods. Qualitative and quantitative comparison of results approved capabilities of the proposed method compared to other selected noise attenuation method.

Passive acoustic identification of bubble flow regime based on synchrosqueezing wavelet transform and deep learning

AbstractA passive acoustic method based on synchrosqueezing wavelet transform (SWT) and deep learning was proposed to automatically identify bubble flow regimes in process industries. The method was established on the bijection relationship between bubble flow regime and the time‐frequency (TF) textures of its passive acoustic emission (PAE) signals. Specifically, the PAE signal of the bubble flow was first acquired by hydrophones, then converted into TF representations (TFRs) by SWT and finally used to train convolutional neural network (CNN) to identify bubble flow regimes automatically. The effects of TF analysis method and CNN architecture were studied: the SWT represented TF textures of PAE signals the best, and its combination with ResNet gave the highest identification accuracy of 99.67%. As its main contribution, the method validated the feasibility of PAE and CNN in bubble flow regime identification, and achieved automatic identification of bubble flow regime with high accuracy and barely no subjective interference.

A Hybrid Approach for Time-Varying Harmonic and Interharmonic Detection Using Synchrosqueezing Wavelet Transform

With widespread non-linear loads and the increasing penetration of distributed generations in the power system, harmonic pollution has become a great concern. The causes of harmonic pollution not only include the integer harmonics, but also interharmonics, which exacerbate the complexity of harmonic analysis. In addition, the output variability of highly non-linear loads and renewables such as electric arc furnaces and photovoltaic solar or wind generation may lead to weakly time-varying harmonics and interharmonics in both frequency and magnitude. These features present challenges for accurate assessment of associated power-quality (PQ) disturbances. To tackle such increasing time-varying PQ problems, a hybrid detection method using synchrosqueezing wavelet transform (SSWT) is proposed. The proposed method first obtains the proper parameter values for the mother wavelet according to numerical computations. The wavelet transform-based synchrosqueezing and a clustering method are applied to determine each frequency component of the waveform under assessment. The time-domain waveform and the associated magnitude of each frequency component is then reconstructed by the inverse SSWT operation. The novelty of the proposed method is that it can decompose the measured waveform containing both harmonics and interharmonics into intrinsic mode functions without the need for fundamental frequency detection. Compared to other time–frequency analysis methods, SSWT has better anti-noise and higher resolution of time–frequency curves; even the measured signal has close frequency components. Simulation results and actual measurement validations show that the proposed method is effective and relatively accurate in time-varying harmonic and interharmonic detection and is suitable for applications in power networks and microgrids that have high penetration of renewables or non-linear loads causing time-varying voltage or current waveforms.

Dual-Core Denoised Synchrosqueezing Wavelet Transform for Gear Fault Detection

Due to the heavy background noise and strong interference, a blurry or even false time-frequency (TF) representation (TFR) by synchrosqueezing wavelet transform (SST) always results in inaccurate or meaningless gear fault extractions. To solve this problem, a novel TF analysis (TFA) method termed dual-core denoised SST is proposed in this article. The concepts critical to the proposed method are: 1) The dual-core denoising is proposed. First, the measured signal is transformed to the binary tree structure of the prewhitening and pseudo-characterizing signals by cepstrum editing. Second, two empirical mode decomposition (EMD)-based denoising techniques are, respectively, proposed for each binary-tree signal, focused on extracting different gear features such as the modulation and transients. Then, the purified feature signal is obtained by mixing the dual-core denoising results. 2) The multistep denoising strategy is studied for further improving the denoising accuracy. The steps are determined by the Rényi entropy. 3) By the multistep dual-core denoising, the denoised TFR with such meaningful TF signatures of gear faults as the accurate demodulation components and distinct transients could be obtained for gear fault detection. The repeatable simulations and engineering applications are used to verify the performance of denoising and TF resolution enhancement for gear fault detection.

An Improved Multi-Ridge Extraction Method Based on Differential Synchro-Squeezing Wavelet Transform

It is very important to accurately extract the instantaneous frequency (IF) of complex non-stationary signals, so the signal processing methods based on IF are widely used in engineering. However, in non-stationary complex vibration signals, it is difficult to extract the instantaneous frequency under strong noise because the frequency changes rapidly. This paper proposes an improved multi-ridge extraction method based on differential synchro-squeezing wavelet transform (DSWT). First, an improved method DSWT is proposed to reduce the computation time and image noise of synchro-squeezing wavelet transform (SWT). The running time of the algorithm can be further reduced by segmenting the signal and compressing the time-frequency matrix. The image noise can be significantly reduced by differentiating SWT time-frequency matrix. It lays a good foundation for the extraction of the time-frequency ridge line. Secondly, an improved multi-ridge extraction method is proposed. The ridge jump due to end effect can be eliminated adaptively in segmented extraction. It further improves the time-frequency aggregation and extraction effect of the wavelet ridgeline image. Finally, the improved multi-ridge extraction method based on DSWT (DSWT-IMRE) is verified by the simulation data and experimental data under different conditions. Compared with the method proposed in previous study, the results show that DSWT-IMRE has higher extraction accuracy.