Time-frequency Characteristics Research Articles

A Study on the Time-Frequency Characteristics of Electromagnetic Radiation Signals during Concrete Cracking

At the moment of rupture, brittle materials such as concrete, coal, and rocks experience a sudden stress change that generates oscillating positive and negative charges on the crack surfaces, resulting in the emission of electromagnetic radiation (EMR) signals. While extensive investigations have focused on the EMR signals produced during the fracturing of coal and rocks, limited attention has been given to the signals arising from concrete cracking. Previous studies have predominantly concentrated on amplitude or frequency domain analysis, lacking a comprehensive exploration of the time-frequency domain characteristics of electromagnetic signals. In this study, various magnetic field sensors capable of collecting frequencies ranging from 50Hz to 50kHz are developed. A laboratory testing platform for concrete cracking is established with the utilization of data acquisition equipment and a rock point load tester. The EMR signals are subjected to time-frequency domain analysis using Short-Time Fourier Transform (STFT) methods. The study explores the relationship between the time-frequency domain characteristics of EMR signals and concrete dimensions, as well as the instantaneous stress drop at the moment of fracture. Experimental findings reveal that the dominant frequency of EMR signals generated during concrete cracking is in the low-frequency range. As the instantaneous stress drop at the moment of fracture increases, the frequency of the EMR signals shifts towards higher frequencies and the signals around 10kHz has a longer duration. When the frequency of EMR signals is lower than 20kHz, larger concrete dimensions result in greater stress drops at the moment of fracture, leading to higher intensity EMR signals. Utilizing the EMR characteristics of concrete cracking holds promise for monitoring the safety and stability of concrete structures.

Bayesian optimization based extreme gradient boosting and GPR time-frequency features for the recognition of moisture damage in asphalt pavement

Moisture damage is one of the major defects in asphalt pavement, and will evolve into potholes in a short time which will affect traffic safety. Ground Penetrating Radar (GPR) is an effective non-destructive testing (NDT) method for detecting moisture damage but its data explanation replies on human experience and subjects to labor intensive. To address this issue, an automatic detection method based on extreme gradient boosting (XGBoost) combined with Bayesian hyper-parameter optimization (BHPO) was proposed to detect moisture damage area from GPR traces. High frequency GPR antenna with 2.3 GHz was used to detect the moisture damage area from simulation, laboratory and field tests, and moisture damage dataset with 7960 traces was collected. Thirty time-frequency parameters were extracted from each GPR trace, normalized to unify the three data source, and then optimized into 12 sensitive parameters by feature importance method. These 12 parameters were used to build the recognition model with XGBoost, and the model tuning parameters were optimized by BHPO. To obtain optimization model, random forest (RF) and artificial neural network (ANN) were also trained with BHPO, and compared with XGBoost model. The results indicate that performance of XGBoost model with BHPO achieves the highest accuracy and lowest time cost both in moisture damage and normal trace classification, the accuracies for moisture damage are XGBoost (96.9%) > ANN (95.6%) > RF (95.4%), respectively, and normal are XGBoost (96.5%) > RF (96.1%) and ANN (96.0%), respectively. On this basis, field tests were conducted by core samples, which verified the correct result of XGBoost model. Our method provides a swift and accurate method to locate subsurface targets directly from GPR signals.

Effective multispike learning in a spiking neural network with a new temporal feedback backpropagation for breast cancer detection

This paper presents an effective learning multi-spike deep spiking neural network with temporal feedback backpropagation for breast cancer detection using contrast-enhanced MRI images. The learning in the spiking network is a new universal temporal feedback called Temporal_Feedback_SpikeProp (TF_SpikeProp) and Temporal_Feedback_ReSuMe (TF_ReSuMe). Thus, it can be implemented on all kinds of algorithms such as MultiSpikeProp and MultiReSuMe algorithms and is compatible with all temporal codings. The presented spiking network is a functional network with high accuracy and convergence in low epochs. The new algorithm explores the influence of all presynaptic neurons on the output error and reduces the role of inactive neurons. Therefore, the error propagation of the spiking network is such that it affects not only the spike time of each neuron in the output spike time, but also the refractory time and the presynaptic spikes of the neuron during the spiking time of the output neuron. The spiking network can correct and adjust the weights, spike delay, spike threshold, and also the time constant of the spike collection kernel. In order to diagnose cancer tissue, time–frequency features such as STFT and packet wavelet transform (WPT) have been used along with texture recognition features such as co-occurrence matrix. The presented network achieved an accuracy of 98.3% in 23 epochs in SpikeProp algorithm, 97.4% in ReSuMe with Duke MRI dataset, 98.64% on MNIST, and 96.1% on Iris. The results have shown that the training algorithm used in this study can achieve high accuracy in low epochs compared to traditional algorithms, while solving the challenge of getting stuck in local minima and poor convergence, as well as reducing the problem of time confusion of output spikes to an acceptable level and providing a fully functional network in practice and reality.

EEG Emotion Recognition Network Based on Attention and Spatiotemporal Convolution

Human emotions are complex psychological and physiological responses to external stimuli. Correctly identifying and providing feedback on emotions is an important goal in human–computer interaction research. Compared to facial expressions, speech, or other physiological signals, using electroencephalogram (EEG) signals for the task of emotion recognition has advantages in terms of authenticity, objectivity, and high reliability; thus, it is attracting increasing attention from researchers. However, the current methods have significant room for improvement in terms of the combination of information exchange between different brain regions and time–frequency feature extraction. Therefore, this paper proposes an EEG emotion recognition network, namely, self-organized graph pesudo-3D convolution (SOGPCN), based on attention and spatiotemporal convolution. Unlike previous methods that directly construct graph structures for brain channels, the proposed SOGPCN method considers that the spatial relationships between electrodes in each frequency band differ. First, a self-organizing map is constructed for each channel in each frequency band to obtain the 10 most relevant channels to the current channel, and graph convolution is employed to capture the spatial relationships between all channels in the self-organizing map constructed for each channel in each frequency band. Then, pseudo-three-dimensional convolution combined with partial dot product attention is implemented to extract the temporal features of the EEG sequence. Finally, LSTM is employed to learn the contextual information between adjacent time-series data. Subject-dependent and subject-independent experiments are conducted on the SEED dataset to evaluate the performance of the proposed SOGPCN method, which achieves recognition accuracies of 95.26% and 94.22%, respectively, indicating that the proposed method outperforms several baseline methods.

Research on Bearing Remaining Useful Life Anti-noise Prediction Based on Fusion of Color-Grayscale Time-Frequency Features

Abstract In contemporary industrial processes, vibration signals collected from bearings often contain significant noise, challenging the efficacy of conventional predictive models in extracting critical degradation features and accurately predicting the remaining useful life (RUL) of bearings. Addressing these challenges, this paper introduces a novel method for predicting bearing RUL under noisy conditions, leveraging a dual-branch multi-scale convolutional attention network (DMCSA) integrated with a dense residual feature fusion network (DRF). Initially, the method applies continuous wavelet trans-form (CWT) to vibration signals to extract color time-frequency image data, followed by grayscale processing to construct a comprehensive color-grayscale time-frequency image dataset, thereby augmenting the model's input features. Enhanced channel and spatial attention mechanisms, combined with multi-scale convolutions, facilitate supe-rior feature extraction and selection. The model's resilience to noise is fortified by in-corporating noise into the training dataset. Subsequently, selected color-gray time-frequency features undergo fusion and relearning through the DRF framework at the model's backend. The crayfish optimization algorithm (COA) is deployed for the astute determination of the model's critical hyperparameters. The proposed DMCSA-DRF model is then applied to predict the health indicator (MSCA-DRF-HI) of the test dataset, culminating in the accurate prediction of the bearings' RUL. Validation experiments demonstrate that our method surpasses comparative models in prediction accuracy un-der diverse noise interferences, signifying a substantial advancement in predictive performance.

Ensemble Machine Learning Approach for Parkinson’s Disease Detection Using Speech Signals

The detection of Parkinson’s disease (PD) is vital as it affects the population worldwide and decreases the quality of life. The disability and death rate due to PD is increasing at an unprecedented rate, more than any other neurological disorder. To this date, no diagnostic procedures exist for this disease. However, several computational approaches have proven successful in detecting PD at early stages, overcoming the disadvantages of traditional methods of diagnosis. In this study, a machine learning (ML) detection system based on the voice signals of PD patients is proposed. The AdaBoost classifier has been utilized to construct the model and trained on a dataset obtained from the machine learning repository of the University of California, Irvine (UCI). This dataset includes voice attributes such as time-frequency features, Mel frequency cepstral coefficients, wavelet transform features, vocal fold features, and tremor waveform quality time. The model demonstrated promising performance, achieving high accuracy, precision, recall, F1 score, and AUC score of 0.96, 0.98, 0.93, 0.95, and 0.99, respectively. Furthermore, the robustness of the proposed model is rigorously assessed through cross-validation, revealing consistent performance across all iterations. The overarching objective of this study is to contribute to the scientific community by furnishing a robust system for the detection of PD.

Mathematical modeling and time-frequency characterization of aero-engine actuators

The actuator is an essential part of the aero-engine control system, and its performance directly determines the efficiency of the whole system. The mathematical model of the aero-engine actuators, including electro-hydraulic servo valves, hydraulic cylinders, and actuators, is modeled and the time-frequency characteristics are analyzed using the Simulink module. Then, the system response is simulated and its time-frequency characteristics are analyzed. The simulation results show that the simulation error of the established actuator model is within the allowable range. The displacement response and the servo valve flow response fluctuate slightly near the steady state value, and the error value is only 0.857% at the maximum fluctuation value, and the system still works normally when a dynamic load is added to the system.

Sound classification with time-frequency features in forest environment

The study of forest sound classification has drawn more attention recently due to its potential for illegal activities and natural disaster monitoring. Based on the forest sound classification dataset (FSC22), a dataset specific to possible sound existing in the forest, five classification methods are utilized to investigate the relationship between recognition accuracy and the number of sound acoustic features, as well as the number of target classes. The results confirmed that extreme random forest is the best method for forest sound classification, with an accuracy of around 70% when the target class number is above 20. Further, Mel-frequency cepstral coefficients are the critical feature for sound classification, while fuzzy labels in the dataset may reduce the success rate of recognition.

Tunnel indirect damage identification algorithm based on service train dynamic response time series

In this study, a tunnel indirect damage identification algorithm based on a service train dynamic response time series was proposed. First, the principles and processes of the proposed algorithm are introduced. The instantaneous frequency and spectral entropy of the trend and seasonal terms of the train dynamic response decomposed by Seasonal and Trend decomposition using Loess are calculated, respectively, to further extract the time-frequency features of the signal. The calculated sequence is input into the Bi-directional Long Short-Term Memory network, and the type of damage case is the output. Second, the feasibility of the algorithm is verified using a model test. Finally, the key links in the algorithm are discussed. The results indicate that the proposed algorithm can identify damage cases with a small number of training and test sets, and the accuracy of the test set can reach 83.33%. Through time-frequency feature extraction and normalization processing, the length of the network input sequence is reduced, the training speed is increased, and damage identification can be realized without a deep network.

Nonlinear vibration and acoustic radiation of an internally resonant buckled beam

Although the nonlinear vibration of buckled beam has been thoroughly studied, its acoustic radiation does not get much attention. This paper deals with the nonlinear vibro-acoustic problem of an internally resonant buckled beam immersed in an infinite fluid. A finite element method based on an arbitrary Lagrangian-Eulerian framework is adopted to analyze the nonlinear interaction between the large-deformed buckled beam and the finite-amplitude acoustic waves in the fluid. The effects of acoustic pressure excitation, external harmonic excitation, and internal resonance on the vibro-acoustic responses of the buckled beam are discussed in the first and second primary resonances. It is found that the amplitude of the acoustic pressure can be comparable to the amplitude of the external harmonic force. This leads to an increase in the critical harmonic excitation amplitude for the occurrence of quasi-periodic responses and an increase in the second mode frequency component of quasi-periodic responses compared to the case neglecting the acoustic pressure excitation on the beam. Saturation phenomena are discovered in the second mode frequency components of beam displacement and acoustic pressure. Due to the 2:1 internal resonance, different time-frequency characteristics, quadrupole acoustic directivity patterns with low radiation efficiency, and dipole acoustic directivity patterns with high radiation efficiency can be observed at different harmonic excitation amplitudes and frequencies. Understanding these phenomena will help design loudspeakers and recognize sound sources.

EEG-Based Mental Workload Classification Method Based on Hybrid Deep Learning Model Under IoT.

Automatically detecting human mental workload to prevent mental diseases is highly important. With the development of information technology, remote detection of mental workload is expected. The development of artificial intelligence and Internet of Things technology will also enable the identification of mental workload remotely based on human physiological signals. In this article, a method based on the spatial and time-frequency domains of electroencephalography (EEG) signals is proposed to improve the classification accuracy of mental workload. Moreover, a hybrid deep learning model is presented. First, the spatial domain features of different brain regions are proposed. Simultaneously, EEG time-frequency domain information is obtained based on wavelet transform. The spatial and time-frequency domain features are input into two types of deep learning models for mental workload classification. To validate the performance of the proposed method, the Simultaneous Task EEG Workload public database is used. Compared with the existing methods, the proposed approach shows higher classification accuracy. It provides a novel means of assessing mental workload.

Research on fault diagnosis method of vehicle cable terminal based on time series segmentation for graph neural network model

In the field of partial discharge (PD) pattern recognition for vehicle cable terminals, the existing recognition methods often lead to reduced accuracy due to inadequate time–frequency features. This study introduces a novel method for time sequence segmentation to construct graph signals. Additionally, we flexibly integrate the graph self-attention convolution layer (GAT) and the self-attention graph pooling layer (SAG) to build a diagnostic model, allowing for robust feature extraction through multiple attention heads of GAT and effective integration of global features via the SAG pooling layer. High-frequency pulse current was utilized for PD testing on four defect models, with subsequent evaluation of outcomes. Furthermore, examinations of cable terminations in real trains provide further support for our approach by improving recognition accuracy and enhancing train operational stability.

Experimental study of multiscale dynamic characterization of the gas-solid phase in a disturbing rotary centrifugal air classifier using pressure signal testing and analysis methods

This study investigates the complex dynamic characteristics of the flow field inside a disturbing rotary centrifugal air classifier through experimental analysis. To this end, micro-pressure sensors are employed to measure pressure signals in different spatial regions under various operating conditions. The analysis includes time-frequency characterization, complexity assessment, and multi-scale energy analysis. The time-domain fluctuation characteristics of the air classifier's pressure signal are closely tied to operating parameters, with the most intense pressure fluctuations occurring in the middle and back region of the air classifier. The loaded condition intensifies the fluctuations, with particle presence exacerbating low-frequency pressure variations and affecting the cyclone's vortex stability. Additionally, the internal installation of the screen cage structure exhibits maximum complexity and randomness in the internal flow field when Rf = 45 Hz, f = 0.4 kg/s, and β = 0°, showing higher randomness and reflecting fine-scale interactions between gas and particles. Conversely, larger scale characteristics are revealed by DET values in higher detail signal levels. Changes in operational parameters, such as perturbation frequency and feeding time, significantly affect the axial distribution of multi-scale energy in the air classifier, with a greater influence on pressure fluctuation at the macroscopic scale, indicating intensified turbulent behavior between gas and particles. This study reveals the influencing mechanisms of various operating parameters on the multiscale dynamic characteristics and complexity of the flow field inside the rotary centrifugal air classifier.

Power quality disturbances classification with imbalanced/insufficient samples based on WGAN-GP-SA and DCNN

The accurate identification of power quality disturbance signals (PQDs) is of great importance to ensure the safe and stable operation of power systems. Considering that there are insufficient samples and imbalance samples of power quality disturbance signals in real systems, the hybrid classification method based on WGAN-GP-SA (Wasserstein generative adversarial networks with gradient penalty and Squeeze-and-Excitation block with atrous spatial pyramid pooling module) and DCNN (deep convolutional neural network) is proposed for recognition of PQDs. Firstly, time-frequency features are extracted by combining improved Kalman and continuous wavelet transforms (CWT); secondly, the original training time-frequency maps are expanded by using WGAN-GP-SA, which can learn the distribution of real data and generate similar samples, and PSNR, SSIM, FID and other indexes are introduced for evaluation of generated data; finally, classification experiments are conducted on two data sets using DCNN, and simulation results show that the accuracy is improved by 2.82 % compared with the original data without augmentation. Furthermore, t-distributed stochastic neighbor embedding algorithm (t-SNE) is introduced for visually analysing of classification performance of various datasets, and the visualization indicates the superiority of the hybrid classification method.

Time-varying damage evolution of concrete under cyclic disturbance loading- an experimental observation utilizing AE and DIC

Cyclic disturbance loading affects the continuous operation of concrete structures. This study aims to better understand the failure behavior of concrete structures under cyclic disturbance loading. This study combined acoustic emission (AE) and digital image correlation (DIC) technologies to study mechanical failure behavior and damage evolution of concrete materials under cyclic disturbance loading and unloading. The AE event time-frequency characteristics, surface principal strain program, and damage parameters of samples with different cyclic disturbance amplitudes were obtained by theoretical calculation. The results reveal that when minor disturbance amplitudes are applied, the elastic modulus of concrete specimens tends to increase. When the disturbance amplitude is too large, the elastic modulus of the specimens drops, leading to faster specimen failure. The AE characteristic parameters exhibit a Felicity effect after the sample is subjected to a cyclic disturbance loading. Moreover, the larger the amplitude of the disturbance load, the more frequent the low-frequency and high-amplitude events generated during the final rupture, the higher the concentration of the principal strain on the sample's surface, and the wider the crack band distribution. AE energy and Effective crack are consistent in characterizing damage, and the changing trend of these two damage parameters can better reflect the deterioration process of concrete. The research results will provide practical and theoretical insights for safely operating and maintaining concrete structures.

A multi-output fault diagnosis framework for hydraulic system using a CNN-SVM hierarchical learning strategy

Achieving asymptotic and concurrent fault diagnosis in hydraulic system remains a challenging endeavor due to the inherent attributes of the hidden occurrence, simultaneous manifestation, coupling, and limited sample size. To address the above issues, this paper proposes a hierarchical multi-output fault detection and diagnosis framework, namely, HMDF, based on a hierarchical learning strategy to leverage an improved convolutional neural network (CNN) and support vector machine (SVM). Both a multi-channel CNN and a multi-branch CNN are employed to extract and downscale features collected by the sensors at diverse sampling frequencies first, and then, such features are subsequently subjected to classification using SVM. The hierarchical learning strategy enables the identification of different fault states, both at the component and the intra-component level. Additionally, a modified whale optimization algorithm is also utilized to optimize the classification process of SVM. Extensive experiments are conducted to test the proposed HMDF with the hydraulic system datasets. Results show that HMDF achieves a diagnostic accuracy of up to 98.9% for the dataset, surpassing traditional methods reliant on manual extraction of time–frequency features, and it also exhibits superior classification performances with a small sample size. The HMDF is expected to offer a generalized framework for the multi-output fault detection and diagnosis in hydraulic systems and other complex components.

Study on time–frequency features of induced charge signals during the damage and failure process of coal medium

Monitoring and preventing coal-rock dynamic disasters are vital for safe mining. To investigate the time–frequency features of induced charge signals with coal damage and failure of roadways, the generation mechanism of free charge in loaded coal is analyzed and the induced charge monitoring test is conducted. According to the stress-induced charge-time curves, the time-domain features of charge signals at each loading stage are obtained. The wavelet threshold denoising approach and generalized Morse wavelet transform method are applied to denoise the raw signals and study the frequency-domain features. Further, the quantitative relationship between the de-noised induced charge signals and the degree of coal damage is established. The results show that the event number, amplitude and fluctuation degree of available induced charge signals are all at a low level in the compaction and elastic stages of the coal, which are mainly generated by the piezoelectric effect and predominantly represent discreteness. When entering the plastic and failure stages, the available signals are primarily produced by the crack propagation and triboelectric effects, with a significant increase in the event number, amplitude, and fluctuation degree. Then the induced charge signals gradually transit from discrete to continuous. Generally, the dominant frequency of the available induced charge signals during the coal damage process is concentrated at 0 ~ 11 Hz. The available induced charge is positively correlated with the degree of coal damage, which can perform the damage degree of coal mass, providing a new approach to evaluate the stability of roadway surrounding rocks.

High-order iterative rearrangement transform for time–frequency characterization of bearing fault impact

Time–frequency analysis (TFA) can effectively characterize features of non-stationary signals. Traditional TFA algorithms construct signal models in the time domain and make the assumption that the instantaneous characteristics of each component are continuous. However, the instantaneous frequency (IF) of the transient signal is discontinuous in the time domain and exhibits a multifaceted relationship with time, such as shock, vibration wave, damped sound wave, etc. Additionally, in most existing TFA methods, low-order group delay (GD) is used to describe transient signals, which leads to unsatisfactory energy concentration and calculation accuracy. To address about issues, a novel TFA technique, termed high-order iterative rearrangement transform (HOIRT), is developed in this research. First, the signal model is defined within the frequency domain, and the frequency ridge of the transient signal is described by a high-order GD (HOGD), which is similar to the IF. Second, a HOGD-based iterative synchrosqueezing operator is defined to reassign time–frequency coefficients into the GD trajectories along the time direction. Finally, the HOGD-based frequency extraction operator is constructed to only retain the target time–frequency information of the transient signal from the rearranged results, such that the noise interference is eliminated and the energy-concentrated TFR is obtained. A simulation signal with nonlinear GDs is employed to illustrate the effectiveness of the HOIRT. Compared with the other seven typical TFA algorithms, the developed technique has the smallest calculation error and Rényi entropy, showing that the HOIRT has the highest accuracy and energy concentration. Analysis result of the bearing fault impact signal shows that the proposed HOIRT can display the time when pulses occur while ensuring high time–frequency resolution, making it suitable for detecting bearing faults.

A Variable-Scale Attention Mechanism Guided Time-Frequency Feature Fusion Transfer Learning Method for Bearing Fault Diagnosis in an Annealing Kiln Roller System

A Variable-Scale Attention Mechanism Guided Time-Frequency Feature Fusion Transfer Learning Method for Bearing Fault Diagnosis in an Annealing Kiln Roller System

Ensemble learning model for concrete delamination depth detection using impact echo

Delamination in concrete structures typically arises from the expansion of steel bar corrosion and the separation of the bonding layer. The presence of these delamination defects leads to a reduction in overall structural integrity, safety, and longevity. However, the delamination depth localization remains challenging. To address this issue, this paper introduces a predictive approach that combines time-frequency features extracted from vibration signals with ensemble learning models. The innovation presented in this paper can be summarized in two key aspects: (1) Signal time-frequency spectrograms are transformed into image features using the Histogram of Oriented Gradients (HOG) method, which serves as input attributes. The utilization of ensemble learning models facilitates both rapid and precise depth estimation of delamination signals, especially those with shallow defects. (2) By analyzing the time-frequency spectrogram of the vibration signals, a double exponential model is established between the depth-to-thickness ratio of delamination and the pixel matrix Eulerian distance. This may provide a new perspective for the depth estimation of delamination defects whose corresponding main frequencies are difficult to locate due to the influence of geometry and boundary effects. Laboratory experiments were conducted on concrete slab with simulated delamination defects at different depths to confirm the proposed method's effectiveness.