Low-frequency Characteristics Research Articles

Wavelet-powered hierarchical frequency filtering framework for autonomous vehicle sensors fault diagnosis and correction under open environments

The reliable operation of autonomous vehicles heavily depends on the functionality of numerous sensors that facilitate environmental perception and vehicle control. However, sensor malfunctions can significantly undermine the dependability and safety of autonomous vehicles. Despite the significant advancements made in sensor fault diagnosis through deep learning, the technology's susceptibility to noise and limited frequency analysis capacity renders it less effective in addressing frequency aliasing and noise interference issues. In addition, repairing incorrect sensor data holds greater significance for autonomous vehicles than merely identifying sensor failures. To this end, this paper proposes a wavelet-powered hierarchical frequency filtering framework (WavePHF) for autonomous vehicle sensors' fault diagnosis and correction. First, this study designs a multi-sensor fault identification and correction joint architecture to detect sensor faults and repair fault signals in time. Second, this study integrates the discrete wavelet transform (DWT) algorithm into the deep model to significantly enhance the model's frequency analysis capabilities and suppress frequency aliasing and noise interference. In particular, DWT is deeply integrated into the convolutional modules' parameter updating and optimization process, thereby facilitating mutual learning and promotion. Third, this study proposes an adaptive frequency weighting block for low-frequency features and an adaptive frequency discarding block for high-frequency features to achieve hierarchical frequency filtering in a coarse-to-fine manner and efficiently identify fault categories and restore pure signals. The WavePHF is evaluated on a real autonomous driving dataset. Experiments show that WavePHF has excellent autonomous vehicle sensors fault diagnosis and correction capabilities and competitive performance in different urban scenarios.

Experimental study on a dual synthetic jets liquid cooling device

To address the growing heat dissipation requirements of high heat flux electronic devices, this work presents experimental research on a dual synthetic jets (DSJ) liquid cooling device. The effects of different heat flux, channel inlet flow rates, driving frequencies, and other parameters on the device ability have been explored. The experimental results indicate that the wall temperature increases by increasing the heat flux. Raising the channel inlet flow rate will reduce wall temperature. When DSJ is off, increasing the inlet flow rate from 0.16 L/min to 0.40 L/min results in a maximum temperature decrease of 6.30 °C and a pressure drop increase of 2.20 kPa. There is essentially no change in the pressure drop when DSJ is operating. When the inlet flow rate is 0.16 L/min, the maximum wall temperature at DSJ on is reduced by 6.83 °C compared with DSJ off. The convective heat transfer coefficient was raised by 40.55%. The vibrating diaphragm exhibits better low-frequency characteristics when working underwater. The convective heat transfer coefficient increases by 36.46% when the driving frequency is 20 Hz. Increasing the driving voltage, the device cooling performance increased. By increasing the driving voltage to 240 V, the wall temperature decreased by 6.30 °C, a decrease of 13.82%, and the convective heat transfer coefficient raised by 28.51%. The device overall performance increased by a maximum of 39.55% after DSJ on compared with DSJ off.

ECG signal fusion reconstruction via hash autoencoder and margin semantic reinforcement

The ECG signal is often accompanied by noise, which can affect its shape characteristics, so it is important to perform signal de-noising. However, the commonly used signal noise reduction methods, such as wavelet or filter transformation, often prioritize high-frequency signals over low-frequency ones, leading to the loss of low-frequency band features or difficulties in capturing them. We propose a fusion reconstruction framework that combines hash autoencoder and margin semantic reinforcement to enhance low-frequency band features. Specifically, for labeled samples, margin semantic reinforcement identifies and corrects weight discrepancies among bands with similar waveforms but different labels to amplify the low-frequency signals associated with the label and reduce irrelevant ones. Meanwhile, hash autoencoder utilizes a semantic hash dictionary to reconstruct the original signal and mitigate noise pollution. For unlabeled samples, the hash autoencoder is utilized to generate pseudo-labels, followed by the reproduction of the aforementioned enhanced reconstruction process. The final step involves weighting the two types of signals, enhanced with margin semantics and hash autoencoder reconstruction, to achieve the reconstruction objective of the original signal, facilitating recognition and detection tasks. Experiments conducted on different classical classifiers demonstrate that the reconstructed ECG signals can significantly improve their performance.

Low-frequency non-reciprocal sound propagation features in thermoacoustic waveguide.

Thermoacoustic waveguides are systems of hollow tubes and thermally graded porous segments that can operate as active materials where acoustic waves receive energy from an external heat source. This work demonstrates that by adjusting the pore geometry several unique low-frequency propagation features arise from the complex-valued band structure of periodic thermoacoustic waveguides that reflect into the acoustic pressure field within finite-length systems. Numerical methods have been employed to model waveguides with porous segments constituted by cylindrical inclusions (parallel pins). In periodic structures, a critical frequency emerges where the sign of the refractive index in one direction of propagation changes, thus zero- and negative-unidirectional refractive index, unidirectional energy transport, and amplification/attenuation crossover effects may take place. On the other hand, the study of the acoustic pressure field shows that, for wave packets with either direction of propagation, finite-length waveguides may behave as active acoustic metamaterials with zero- or negative-refractive index. The acoustic pressure field in the waveguide, generated by an upstream source, may exhibit increasing amplitude and phase recovery farther away from the source, mimicking the field created by a downstream source, propagating upstream in a non-active medium.

Electro-mechano transduction of stacked triboelectric nanogenerator considering wave force: Modeling, validation and applications

Aiming at the low-frequency characteristics of ocean wave energy, triboelectric nanogenerators (TENGs) provide an efficient collection method. We propose a multi-layer contact-separation mode triboelectric nanogenerator (MS-TENG) based on a frequency-up mechanism to collect water wave energy. First, we study the effects of frequency rising and stacking numbers of TENG on the output of MS-TENG. Experiments show that the short-circuit current of MS-TENG with 12 layers stacked is 152.16 μA when the rotating speed is 60 rpm. Compared with a single TENG unit, the short-circuit current of MS-TENG is increased by 6.35 times. Second, the hydrodynamic simulation method is used to study the force and energy capture efficiency of floating bodies with different shapes in water waves. The results show that the energy capture efficiency of the cylindrical floating body is 70.14 % higher than that of the spherical floating body under the same volume. Third, we use a cylindrical floating body to package the MS-TENG. In water tank experiments, when the load resistance is 4 MΩ, the peak power density of the packaged MS-TENG is 14.56 W/m3. In addition, with the help of MS-TENG, we set up a wireless temperature and humidity monitoring system to realize the preliminary marine environmental monitoring.

Aerodynamic Load Variation and Wind-Induced Vibration Analysis of Tower Cranes Under Full Wind Angles

When tower cranes are applied in coastal engineering, the wind-induced vibration under complex wind loads along the coast will be exceptionally obvious. It is increasingly important to explore the wind-induced vibration characteristics of different sections of tower cranes under full wind angles. This paper first establishes a three-dimensional numerical wind tunnel model that can be used for the simulation of tower cranes under full wind angles and analyzes the aerodynamic load variation of the tower crane under full wind angles. Secondly, the tower crane structure is divided into sub-regions, and the AR model method in the linear filtering method is used to simulate the three-dimensional wind speed time history. Further, the wind speed time history is applied to the finite element transient analysis, and based on the Newmark-beta method, the acceleration and displacement responses of different sections of the tower crane structure under fluctuating wind are systematically studied. Finally, the equivalent static wind load of different sections of the tower crane is evaluated by using the gust load factor method. The results indicate that the wind load of the tower crane exhibits an obvious three-dimensional effect, and the crosswind load cannot be ignored in the analysis of wind stability. The different sections of the tower crane structure show low-frequency vibration characteristics, and the low-frequency band is dominated by down-wind vibration, while the high-frequency band often experiences crosswind vibration. The amplification effect of the tower crane on fluctuating wind loads will increase with the increase of the height of the tower crane and the length of the jib, and the amplification effect on fluctuating wind loads in the crosswind direction is more obvious. The equivalent static wind load on the top of the tower crane reaches the maximum when the angle between the jib and the incoming wind is 60°.

Low-frequency ERK and Akt activity dynamics are predictive of stochastic cell division events

Understanding the dynamics of intracellular signaling pathways, such as ERK1/2 (ERK) and Akt1/2 (Akt), in the context of cell fate decisions is important for advancing our knowledge of cellular processes and diseases, particularly cancer. While previous studies have established associations between ERK and Akt activities and proliferative cell fate, the heterogeneity of single-cell responses adds complexity to this understanding. This study employed a data-driven approach to address this challenge, developing machine learning models trained on a dataset of growth factor-induced ERK and Akt activity time courses in single cells, to predict cell division events. The most predictive models were developed by applying discrete wavelet transforms (DWTs) to extract low-frequency features from the time courses, followed by using Ensemble Integration, a data integration and predictive modeling framework. The results demonstrated that these models effectively predicted cell division events in MCF10A cells (F-measure=0.524, AUC=0.726). ERK dynamics were found to be more predictive than Akt, but the combination of both measurements further enhanced predictive performance. The ERK model`s performance also generalized to predicting division events in RPE cells, indicating the potential applicability of these models and our data-driven methodology for predicting cell division across different biological contexts. Interpretation of these models suggested that ERK dynamics throughout the cell cycle, rather than immediately after growth factor stimulation, were associated with the likelihood of cell division. Overall, this work contributes insights into the predictive power of intra-cellular signaling dynamics for cell fate decisions, and highlights the potential of machine learning approaches in unraveling complex cellular behaviors.

Wavelet–pixel domain progressive fusion network for underwater image enhancement

Due to the complexity of the underwater environment, underwater images suffer from low-illumination, color deviation, and blurred details. Conventional underwater image enhancement (UIE) approaches have achieved remarkable success in improving color and illumination in the pixel domain. However, they ignore fine-grained details during the enhancement process. To solve these problems, we propose a novel wavelet–pixel domain progressive fusion network (WPFNet) to improve details while enhancing illumination and reducing color deviation for degraded underwater images. Unlike the UIE method, we propose a wavelet domain module (WDM) to obtain different scale frequency features with fine-grained details. Concretely, we develop a residual attention block (RAB) that optimizes low-frequency sub-images to maintain rich details, and enhance color and illumination information for low-frequency features. Transformer block (T_Block) is proposed to convert three high-frequency sub-images into high-frequency features operated by self-attention mechanism. Frequency features are obtained by adding low- and high-frequency features from the same scale, suffering from insufficient color and illumination information. Thus, the pixel domain module (PDM) is introduced to extract spatial features of different scales with rich color and illumination information in the pixel domain. Reconstruction module (REM) is proposed to fuse frequency and spatial features from small to large scale to restructure clear underwater images. The dual-domain fusion block (DFB) is proposed as a key element of REM, which exploits the constraint signal to fuse superior information of frequency and spatial features. Finally, our WPFNet obtains high-quality underwater images in four benchmark datasets and achieves superior performance compared to state-of-the-art UIE methods. The code is available at: https://github.com/LiuShiBen/WPFNet.

Ship-Radiated Noise Separation in Underwater Acoustic Environments Using a Deep Time-Domain Network

Ship-radiated noise separation is critical in both military and economic domains. However, due to the complex underwater environments with multiple noise sources and reverberation, separating ship-radiated noise poses a significant challenge. Traditionally, underwater acoustic signal separation has employed blind source separation methods based on independent component analysis. Recently, the separation of underwater acoustic signals has been approached as a deep learning problem. This involves learning the features of ship-radiated noise from training data. This paper introduces a deep time-domain network for ship-radiated noise separation by leveraging the power of parallel dilated convolution and group convolution. The separation layer employs parallel dilated convolution operations with varying expansion factors to better extract low-frequency features from the signal envelope while preserving detailed information. Additionally, we use group convolution to reduce the expansion of network size caused by parallel convolution operations, enabling the network to maintain a smaller size and computational complexity while achieving good separation performance. The proposed approach is demonstrated to be superior to the other common networks in the DeepShip dataset through comprehensive comparisons.

Salient object detection in low-light RGB-T scene via spatial-frequency cues mining

Low-light conditions pose significant challenges to vision tasks, such as salient object detection (SOD), due to insufficient photons. Light-insensitive RGB-T SOD models mitigate the above problems to some extent, but they are limited in performance as they only focus on spatial feature fusion while ignoring the frequency discrepancy. To this end, we propose an RGB-T SOD model by mining spatial-frequency cues, called SFMNet, for low-light scenes. Our SFMNet consists of spatial-frequency feature exploration (SFFE) modules and spatial-frequency feature interaction (SFFI) modules. To be specific, the SFFE module aims to separate spatial-frequency features and adaptively extract high and low-frequency features. Moreover, the SFFI module integrates cross-modality and cross-domain information to capture effective feature representations. By deploying both modules in a top-down pathway, our method generates high-quality saliency predictions. Furthermore, we construct the first low-light RGB-T SOD dataset as a benchmark for evaluating performance. Extensive experiments demonstrate that our SFMNet can achieve higher accuracy than the existing models for low-light scenes.

DDFNet-A: Attention-Based Dual-Branch Feature Decomposition Fusion Network for Infrared and Visible Image Fusion

The fusion of infrared and visible images aims to leverage the strengths of both modalities, thereby generating fused images with enhanced visible perception and discrimination capabilities. However, current image fusion methods frequently treat common features between modalities (modality-commonality) and unique features from each modality (modality-distinctiveness) equally during processing, neglecting their distinct characteristics. Therefore, we propose a DDFNet-A for infrared and visible image fusion. DDFNet-A addresses this limitation by decomposing infrared and visible input images into low-frequency features depicting modality-commonality and high-frequency features representing modality-distinctiveness. The extracted low and high features were then fused using distinct methods. In particular, we propose a hybrid attention block (HAB) to improve high-frequency feature extraction ability and a base feature fusion (BFF) module to enhance low-frequency feature fusion ability. Experiments were conducted on public infrared and visible image fusion datasets MSRS, TNO, and VIFB to validate the performance of the proposed network. DDFNet-A achieved competitive results on three datasets, with EN, MI, VIFF, QAB/F, FMI, and Qs metrics reaching the best performance on the TNO dataset, achieving 7.1217, 2.1620, 0.7739, 0.5426, 0.8129, and 0.9079, respectively. These values are 2.06%, 11.95%, 21.04%, 21.52%, 1.04%, and 0.09% higher than those of the second-best methods, respectively. The experimental results confirm that our DDFNet-A achieves better fusion performance than state-of-the-art (SOTA) methods.

Frequency-Separated Attention Network for Image Super-Resolution

The use of deep convolutional neural networks has significantly improved the performance of super-resolution. Employing deeper networks to enhance the non-linear mapping capability from low-resolution (LR) to high-resolution (HR) images has inadvertently weakened the information flow and disrupted long-term memory. Moreover, overly deep networks are challenging to train, thus failing to exhibit the expressive capability commensurate with their depth. High-frequency and low-frequency features in images play different roles in image super-resolution. Networks based on CNNs, which should focus more on high-frequency features, treat these two types of features equally. This results in redundant computations when processing low-frequency features and causes complex and detailed parts of the reconstructed images to appear as smooth as the background. To maintain long-term memory and focus more on the restoration of image details in networks with strong representational capabilities, we propose the Frequency-Separated Attention Network (FSANet), where dense connections ensure the full utilization of multi-level features. In the Feature Extraction Module (FEM), the use of the Res ASPP Module expands the network’s receptive field without increasing its depth. To differentiate between high-frequency and low-frequency features within the network, we introduce the Feature-Separated Attention Block (FSAB). Furthermore, to enhance the quality of the restored images using heuristic features, we incorporate attention mechanisms into the Low-Frequency Attention Block (LFAB) and the High-Frequency Attention Block (HFAB) for processing low-frequency and high-frequency features, respectively. The proposed network outperforms the current state-of-the-art methods in tests on benchmark datasets.

A broadband hybrid blue energy nanogenerator for smart ocean IoT network

Blue energy represents a crucial renewable energy resource; however, its development has been limited by the low and variable frequency characteristics of sea waves. In this study, we present a novel approach, wherein a spring pendulum coupled hybrid energy harvester (SPC-HEH) is employed to address these limitations. The SPC-HEH integrates two electromagnetic generators (EMGs), two piezoelectric nanogenerators (PENGs), and two multilayer-structured triboelectric nanogenerators (TENGs) into a vessel, enabling efficient wave energy harvesting. Notably, the integration of the lightweight PENG and TENG with the heavy EMG not only enhances the water wave energy harvesting capability of SPC-HEH but also optimizes its spatial utilization efficiency. Additionally, owing to the spring pendulum, the SPC-HEH is capable of simultaneously harnessing the wave kinetic energy of horizontal and vertical directions and realize the decoupling of vertical energy capture from horizontal energy capture. Remarkably, owing to its precise geometric design and efficient space utilization, the SPC-HEH achieves an impressive peak power density of 82.4 Wm−3. Finally, we demonstrate the practical applicability of the SPC-HEH by integrating it with a circuit as a self-powered ocean buoy for real-time monitoring of ocean temperature, wave frequency, ocean turbidity and other information. The remarkable potential of this system for integration into self-powered smart ocean IoT networks holds significant promise for advancing marine research and environmental monitoring capabilities.

A joint feature refining and multi-resolution feature learning approach for bandwidth extension of speech coded by EVS codec

The bandwidth of the speech signals is often limited in speech communications due to the specifications of the standardized codecs or insufficient bitrates. So, we present a bandwidth extension (BWE) framework of speech signals coded by the enhanced voice services (EVS) codec. Previous studies on speech bandwidth extension based on convolutional neural network (CNN) which exist channel and spatial redundancy feature information. This work proposes an end-to-end architecture which combines novel channel and spatial reconstruction module. Specifically, the spatial reconstruction module utilizes a mask to distinguish high-frequency and low-frequency features. Group convolutions is then employed to enhance high-frequency features and suppress low-frequency features. The channel reconstruction module is introduced to reduce unnecessary feature information. Additionally, we introduce a novel time–frequency loss function, incorporating time-domain loss and frequency-domain loss based on Mel-spectrum and multi-resolution representation, to optimize the network. To assess the model and loss function’s performance, we conducted experiments on subdatasets encoded at rates of 6.6kbps, 7.2kbps, and 8kbps, respectively. The experimental results demonstrated that our proposed model surpassed other baseline models in terms of LSD, SNR, PESQ, and MOS scores.

Active feedback control on acoustic radiation of metastructure shell with low-frequency and broadband characteristics

Due to the tunable characteristics of elastic waves, the vibroacoustic coupling behavior of a mechanical metastructure is a hot topic of underwater vehicles. In this work, a metastructure shell with active feedback control is presented and fabricated. The dynamic effective density and sound pressure level are derived to find the influences of acceleration and displacement feedback control. Different from a single cylinder, a double cylinder structure has both in-phase and anti-phase modes. Numerical results are obtained by Fourier transform and harmonic series expansion. With the introduction of an active feedback control system, the reduction of acoustic radiation shows low-frequency and broadband characteristics. In addition, finite element simulation is applied to support numerical results and present vibroacoustic characteristics. Finally, an experiment is performed in the anechoic chamber to illustrate the quiet metastructure shell, which can be applied to new designs of underwater vehicles.

DF3Net: Dual frequency feature fusion network with hierarchical transformer for image inpainting

In recent years, researchers have made significant strides in computer vision by leveraging transformers, achieving remarkable breakthroughs in low-level vision tasks. The inherent long-range dependency of transformers grants them potent remote modeling capabilities, surpassing those of Convolutional Neural Networks (CNNs) and enabling the extraction of global features and accurate semantic structures. However, it has been observed that single transformer frameworks lack sensitivity to high-frequency information in images, resulting in the generation of blurry reconstructed regions. To address this limitation, this paper proposes a novel two-branch Dual Frequency Feature Fusion Network (DF3Net) for image inpainting based on the hierarchical atrous transformer (HAT). Specifically, the head of the dual-frequency convolution (DFC) module decouples the feature maps into low and high-frequency components. The low-frequency factor goes through the proposed HAT branch, while the high-frequency component is input into the gated convolution module branch, effectively capturing both global structural information and local texture details. Finally, the DFC tail fuses the high and low-frequency features to output the reconstructed image. Moreover, the feature fusion between high and low-frequency branches is performed layer-wise within the network, enabling mutual learning between the two branches interactively, ensuring coherence in image semantic structure and texture details. Experimental evaluations on Places2, Paris StreetView, and CelebA-HQ with different mask ratios demonstrate that the proposed method outperforms state-of-the-art methods in enhancing the structural accuracy of image inpainting. It generates semantically reasonable images with fine texture details.

Hollow Fe3O4/Fe@C nanocubes for broadband microwave absorption spanning low- and high-frequency bands

Rationally constructing hollow structures with heterogeneous interface is usually an effective strategy to regulate the electromagnetic response behavior for lightweight and strong absorption, but achieving low-frequency and broadband absorption is a major challenge. This work adopts a novel Fe(II)/Fe(III) ions regulation strategy to design and synthesize hollow Fe(OH)X (x = 2, 3) nanocubes, which addresses the disadvantages of time-consuming or using etchant. Subsequently, a unique hollow Fe3O4/Fe@C nanoboxes with adjustable magnetic composition and uniformly coated shell have been successfully synthesized by in-situ carbonization with the assistance of phenolic resin and hydrogen thermal reduction. Notably, the different magnetic components induce the magnetic coupling assembly and play a crucial role in determining its electromagnetic parameters, whereas Fe3O4 contributes to more dipole polarization and stronger low-frequency response characteristics. For Fe3O4@C hollow nanoboxes, the minimum reflection loss (RLmin) is as low as –119.9 dB at 2.3 mm, and extremely wide effective absorption bandwidth (EAB) of 11.4 GHz is achieved spanning low-frequency band of 2.8 ∼ 7.2 GHz and high-frequency band of 11.0 ∼ 18 GHz at 5.0 mm. Surprisingly, the Fe3O4@C also has satisfactory EAB20 (RL＜ –20 dB) reaching 7.1 GHz (14.9 ∼ 18 GHz) at a thin thickness of 1.8 mm. The RLmin of other hollow nanoboxes is also below –30 dB, and the EAB is greater than 9 GHz. This study provides a competitive candidate for efficient microwave absorption in complex frequency bands, especially low-frequency bands.

A Comprehensive Analysis of Trapezius Muscle EMG Activity in Relation to Stress and Meditation

Introduction: This study analyzes the efficacy of trapezius muscle electromyography (EMG) in discerning mental states, namely stress and meditation. Methods: Fifteen healthy participants were monitored to assess their physiological responses to mental stressors and meditation. Sensors were affixed to both the right and left trapezius muscles to capture EMG signals, while simultaneous electroencephalography (EEG) was conducted to validate cognitive states. Results: Our analysis of various EMG features, considering frequency ranges and sensor positioning, revealed significant changes in trapezius muscle activity during stress and meditation. Notably, low-frequency EMG features facilitated enhanced stress detection. For accurate stress identification, sensor configurations can be limited to the right trapezius muscle. Furthermore, the introduction of a novel method for determining asymmetry in EMG features suggests that applying sensors on bilateral trapezius muscles can improve the detection of mental states. Conclusion: This research presents a promising avenue for efficient cognitive state monitoring through compact and convenient sensing.

Transmission tower bolt-loosening time–frequency analysis and localization method considering time-varying characteristics

To address the issues of high concealment and difficult positioning of loose bolts in transmission towers, this paper proposes a new method for locating loose bolts in transmission towers. In this method, we divide the vibration response of the transmission tower into low-frequency signals of 2–25 Hz and high-frequency signals of 25–75 Hz. For the low-frequency signals, the single signal component is obtained by adaptive Chirp mode decomposition and uses the general demodulation transformation to deeply denoise the non-modal information. Since frequency characteristics themselves do not contain time information, considering the importance of time information for positioning, we propose a low-frequency time-varying frequency feature that preserves time characteristics based on synchronous wavelet transform and peak search. For the high-frequency signals, we use singular value decomposition to remove signal outliers caused by pulse excitation and eliminate forced vibrations through wavelet packet transform. Without altering its inherent characteristics, this method enables high-frequency time-domain signals to better represent the nonlinear characteristics of transmission towers. Furthermore, based on the powerful capabilities of Timesnet and Transformer in dealing with time series data, we propose a fault diagnosis model, which ultimately achieves the positioning of loose bolts in transmission towers. The bolt node model proves that this approach can better represent the loose bolt characteristics, and the transmission tower model verifies the effectiveness of this approach in locating loose bolts in transmission towers. Finally, bolt-loosening tests were conducted on a 110 kV transmission tower, and the accuracy of the positioning results reached 92.8%, demonstrating the effectiveness and efficiency of this method in practical positioning applications.

The multi-dimensional joint response of the electric potential time–frequency-field in the fracture process of flawed coal under dynamic load

The fracture failure of flawed coal has always been a problem in underground engineering. Therefore, it is necessary to monitor crack propagation and damage evolution during coal fracture. In this study, the split Hopkinson pressure bar (SHPB) and digital image correlation (DIC) were combined to analyze the impact failure electric potential (EP) of flawed coal samples. The multi-dimensional joint response characteristics of coal EP time–frequency-field under dynamic load were studied and the differential expression and multi-fractal characteristics of EP responses at different measuring points were analyzed. Based on the EP signals, the evolution of damage to the coal was calculated. The results show that the time series evolution of EP and stress–strain curves during coal failure is consistent. The generation of fracture zones will excite the high-value area of local EP, and the spectral distribution of EP signals in this area shows the characteristics of low frequency and high amplitude. Each channel’s EP signal and the degree of deformation at the corresponding place are perfectly matched. The multi-fractal spectrum of the EP signal close to the fracture zone has a higher Δα and a lower Δf, which demonstrates the dominance of the large-scale EP signal produced in this region by the destruction of sample. Theoretical study indicates that the main causes of the anomalous distribution of the EP field are crack tip potential, diffusion potential, etc. The results can provide theoretical basis for the evaluation of coal fracturing by EP technology.