Search

Gearboxes operate in challenging environments, which leads to a heightened incidence of failures, and ambient noise further compromises the accuracy of fault diagnosis. To address this issue, we introduce a fault diagnosis method that employs singular value decomposition (SVD) and graph Fourier transform (GFT). Singular values, commonly employed in feature extraction and fault diagnosis, effectively encapsulate various fault states of mechanical equipment. However, prior methods neglect the inter-relationships among singular values, resulting in the loss of subtle fault information concealed within. To precisely and effectively extract subtle fault information from gear vibration signals, this study incorporates graph signal processing (GSP) technology. Following SVD of the original vibration signal, the method constructs a graph signal using singular values as inputs, enabling the capture of topological relationships among these values and the extraction of concealed fault information. Subsequently, the graph signal undergoes a transformation via GFT, facilitating the extraction of fault features from the graph spectral domain. Ultimately, by assessing the Mahalanobis distance between training and testing samples, distinct defect states are discerned and diagnosed. Experimental results on bearing and gear faults demonstrate that the proposed method exhibits enhanced robustness to noise, enabling accurate and effective diagnosis of gearbox faults in environments with substantial noise. [ABSTRACT FROM AUTHOR]

State-of-the-art smart cities have been calling for economic but efficient energy management over a large-scale network, especially for the electric power system. It is a critical issue to monitor, analyze, and control electric loads of all users in the system. In this study, a non-intrusive load monitoring method was designed for smart power management using computer vision techniques popular in artificial intelligence. First of all, one-dimensional current signals are mapped onto two-dimensional color feature images using signal transforms (including the wavelet transform and discrete Fourier transform) and Gramian Angular Field (GAF) methods. Second, a deep neural network with multi-scale feature extraction and attention mechanism is proposed to recognize all electrical loads from the color feature images. Third, a cloud-based approach was designed for the non-intrusive monitoring of all users, thereby saving energy costs during power system control. Experimental results on both public and private datasets demonstrate that the method achieves superior performances compared to its peers, and thus supports efficient energy management over a large-scale Internet of Things network. [ABSTRACT FROM AUTHOR]

Information on boll distribution within a cotton plant is critical to evaluate the adaptation and response of cotton plants to environmental and biotic stress in cotton production. Cotton researchers have applied available conventional fiber measurements, such as the high volume instrument (HVI) and advanced fiber information system (AFIS), to map the location and the timing of boll development and distribution within plants and further to determine within-plant variability of cotton fiber properties. Both HVI and AFIS require numerous cotton bolls combined for the measurement. As an alternative approach, attenuated total reflection Fourier transform infrared (ATR FT-IR) spectroscopy was proposed to measure fiber maturity (MIR) and crystallinity (CIIR) of a sample as little as 0.5 mg lint. Extending fiber maturity and crystallinity measurement into a single boll for node-by-node mapping, FT-IR method might be advantageous due to less sampling amount compared with HVI and AFIS methods. Results showed that FT-IR technique enabled the evaluation of fiber MIR and CIIR at a boll level, which resulted in average MIR and CIIR values highly correlated with HVI micronaire (MIC) and AFIS maturity ratio (M). Hence, FT-IR technique possesses a good potential for a rapid and non-destructive node-by-node mapping of cotton boll maturity and crystallinity distribution. [ABSTRACT FROM AUTHOR]

Spectrally resolved interferometry utilizing a femtosecond laser is widely employed for absolute distance measurement. However, deviations in the output time pulse of the conventional algorithm through inverse Fourier transform are inevitable. Herein, an improved data processing algorithm employing a time-shifting parameter is proposed to improve the accuracy of spectrally resolved interferometry. The principle of the proposed time-shifting algorithm is analyzed theoretically after clarifying the deviation source of the conventional algorithm. Simulation and experimental work were conducted to indicate the improvement in the accuracy of the output absolute distance. The results demonstrated that the proposed algorithm could reduce the deviation of output distances towards the reference values, reaching 0.58 μm by half compared to the conventional algorithm. Furthermore, the measurement uncertainty was evaluated using the Guide to the Expression of Uncertainty in Measurement (GUM), resulting in an expanded uncertainty of 0.71 μm with a 95% confidence. [ABSTRACT FROM AUTHOR]

This study presents a new approach to determining the preload force of bolted joints. The concept involves measuring the torsional angle without contact. For this purpose, we present a circular magnetic sensor array integrated into the torque wrench. The torsional angle in bolted joints depends on the dimensions of the screw and the materials used and is typically less than four degrees. For this reason, one requirement is a high angular resolution so that a continuous recording of the torsion angle is feasible during the assembly process. This can be achieved using the circular sensor array and adapted signal processing methods. Two signal processing approaches are utilized. First, the direct method uses the discrete Fourier transformation to calculate the rotation angle from the signal phase. This approach is robust to signal distortion and does not depend on signal amplitude. Second, the method with a learning phase employs Gaussian process regression to minimize the angle error. In an experiment, both approaches were applied within a test bench and showed promising results. The direct method demonstrated a very good angular resolution without training and calibration. For mobile and less-complex applications where a reference system is unavailable, the direct method is preferable. However, in complex measurement systems where reference systems can be utilized initially, significant enhancements to an excellent resolution can be achieved through prior training. [ABSTRACT FROM AUTHOR]

Induction motors (IM) play a fundamental role in the industrial sector because they are robust, efficient, and low-cost machines. Changes in the environment, installation errors, or modifications to working conditions can generate faults in induction motors. The trend on IM fault detection is focused on the design techniques and sensors capable of evaluating multiple faults with various signals using non-invasive analysis. The methodology is based on processing electric current signals by applying the short-time Fourier transform (STFT). Additionally, the computation of the mean and standard deviation of infrared thermograms is proposed as main indicators. The proposed system combines both parameters by means of Support Vector Machine and k-nearest-neighbor classifiers. The development of the diagnostic system was done with digital hardware implementations using a Xilinx PYNQ Z2 card that integrates an FPGA with a microprocessor, thus taking advantage of the acquisition and processing of digital signals and images in hardware. The proposed method has proved to be effective for the classification of healthy (HLT), misalignment (MAMT), unbalance (UNB), damaged bearing (BDF), and broken rotor bar (BRB) faults with an accuracy close to 99%. [ABSTRACT FROM AUTHOR]

Long-term coherent integration (CI) can effectively improve the radar detection capability for high-speed targets. However, the range walk (RW) effect caused by high-speed motion significantly degrades the detection performance. To improve detection performance, this study proposes an improved algorithm based on the modified Radon inverse Fourier transform (denoted as IMRIFT). The proposed algorithm uses parameter searching for velocity estimation, designs a compensation function based on the relationship between velocity and distance walk and Doppler ambiguity terms, and performs CI based on the compensated signal. IMRIFT can achieve RW correction, avoid the blind-speed sidelobe (BSSL) effect caused by velocity mismatch, and improve detection performance, while ensuring low computational complexity. In addition, considering the relationship between energy concentration regions and bandwidth in the 2D frequency domain, a fast method based on IMIRFT is proposed, which can balance computational cost and detection capacity. Finally, a series of comparative experiments are conducted to demonstrate the effectiveness of the proposed algorithm and the fast method. [ABSTRACT FROM AUTHOR]

Activity recognition is one of the significant technologies accompanying the development of the Internet of Things (IoT). It can help in recording daily life activities or reporting emergencies, thus improving the user's quality of life and safety, and even easing the workload of caregivers. This study proposes a human activity recognition (HAR) system based on activity data obtained via the micro-Doppler effect, combining a two-stream one-dimensional convolutional neural network (1D-CNN) with a bidirectional gated recurrent unit (BiGRU). Initially, radar sensor data are used to generate information related to time and frequency responses using short-time Fourier transform (STFT). Subsequently, the magnitudes and phase values are calculated and fed into the 1D-CNN and Bi-GRU models to extract spatial and temporal features for subsequent model training and activity recognition. Additionally, we propose a simple cross-channel operation (CCO) to facilitate the exchange of magnitude and phase features between parallel convolutional layers. An open dataset collected through radar, named Rad-HAR, is employed for model training and performance evaluation. Experimental results demonstrate that the proposed 1D-CNN+CCO-BiGRU model demonstrated superior performance, achieving an impressive accuracy rate of 98.2%. This outperformance of existing systems with the radar sensor underscores the proposed model's potential applicability in real-world scenarios, marking a significant advancement in the field of HAR within the IoT framework. [ABSTRACT FROM AUTHOR]

Dispersive interferometry based on a femtosecond laser is extensively utilized for achieving absolute distance measurements with high accuracy. However, this method cannot measure arbitrary distances without encountering a dead zone, and deviations in its output results are inevitable due to inherent theory limitations. Therefore, two enhanced data-processing algorithms are proposed to improve the accuracy and reduce the dead zone of dispersive interferometry. The principles of the two proposed algorithms, namely the truncated-spectrum algorithm and the high-order-angle algorithm, are proposed after explaining the limitations of conventional methods. A series of simulations were conducted on these algorithms to show the improved accuracy of measurement results and the elimination of the dead zone. Furthermore, an experimental setup based on a dispersive interferometer was established for the application of these proposed algorithms to the experimental interference spectral signals. The results demonstrated that compared with the conventional algorithm, the proposed truncated-spectrum algorithm could reduce the output distance deviations derived from direct inverse Fourier transforming by eight times to reach as low as 1.3 μm. Moreover, the unmeasurable dead zone close to the zero position of the conventional algorithm, i.e., the minimum working distance of a dispersive interferometer, could be shortened to 22 μm with the implementation of the proposed high-order-angle algorithm. [ABSTRACT FROM AUTHOR]

The conventional approach to phase reconstruction in Reflective Ghost Imaging (RGI) typically involves the introduction of three reference screens into the reference path, deeming the Fourier transform step indispensable. However, this method introduces complexity to the system and raises concerns regarding potential errors in phase retrieval. In response to these challenges, we advocate for adopting the Five-Step Phase-Shifting (FSPS) method in the RGI system. This method presents two key advantages over traditional approaches: (1) It streamlines the phase reconstruction process by eliminating the requirement for a Fourier inverse transform. (2) It avoids the need to insert objects into the reference optical path, simplifying the computation of reference optical path intensity and enabling seamless application to Computational Ghost Imaging (CGI), overcoming the constraints of Dual-Arm Ghost Imaging (DAGI). We substantiate the theoretical proposition through numerical simulations involving two intricate objects. Furthermore, our discussion delves into exploring the influence of varying reflective angles on the phase reconstruction performance. [ABSTRACT FROM AUTHOR]

An accurate estimation of the time difference of arrival (TDOA) is crucial in localization, communication, and navigation. However, a low signal-to-noise ratio (SNR) can decrease the reliability of the TDOA estimation result. Therefore, this study aims to improve the performance of the TDOA estimation of dual-channel sensors for single-sound sources in low-SNR environments. This study introduces the theory of time rearrangement synchrosqueezing transform (TRST) into the time difference of arrival estimation. While the background noise TF points show random time delays, the signal time-frequency (TF) points originating from uniform directions that exhibit identical lags are considered in this study. In addition, the time difference rearrangement synchrosqueezing transform (TDST) algorithm is developed to separate the signal from the background noise by exploiting its distinct time delay characteristics. The implementation process of the proposed algorithm includes four main steps. First, a rough estimation of the time delay is performed by calculating the partial derivative of the short-time cross-power spectrum. Second, a rearrangement operation is conducted to separate the TF points of the signal and noise. Third, the TF points on both sides of the time-delay energy ridge are extracted. Finally, a refined TDOA estimation is realized by applying the inverse Fourier transformation on the extracted TF points. Furthermore, a second-order-based time difference reassigned synchrosqueezing transform algorithm is proposed to improve the robustness of the TDOA estimation by enhancing the TF energy aggregation. The proposed algorithms are verified by simulations and experiments. The results show that the proposed algorithms are more robust and accurate than the existing algorithms. [ABSTRACT FROM AUTHOR]

Current ghost imaging phase reconstruction schemes require either complex optical systems, iterative algorithms, Fourier transform steps, or entangled photon pairs. These factors may increase the difficulty of system design, lead to phase retrieval errors, or result in excessive time consumption. To tackle this challenge, we propose a five-step phase-shifting method that eliminates the need for complex optical systems, Fourier transform steps, entangled photon pairs, or iterative algorithms. Using five specifically designed incoherent sources, we can generate five distinct ghost imaging patterns. Subsequently, the phase information of the object can be calculated from these five speckle patterns. Additionally, we offer a detailed theoretical explanation for choosing the five-step phase-shifting method over the more commonly used three-step or four-step phase-shifting methods. We demonstrate the applicability of this theoretical proposal through numerical simulations involving two types of complicated objects. The results illustrate that the phase information of the complex object can be successfully and quantitatively reconstructed. [ABSTRACT FROM AUTHOR]

Recent successes in deep learning have inspired researchers to apply deep neural networks to Acoustic Event Classification (AEC). While deep learning methods can train effective AEC models, they are susceptible to overfitting due to the models' high complexity. In this paper, we introduce EnViTSA, an innovative approach that tackles key challenges in AEC. EnViTSA combines an ensemble of Vision Transformers with SpecAugment, a novel data augmentation technique, to significantly enhance AEC performance. Raw acoustic signals are transformed into Log Mel-spectrograms using Short-Time Fourier Transform, resulting in a fixed-size spectrogram representation. To address data scarcity and overfitting issues, we employ SpecAugment to generate additional training samples through time masking and frequency masking. The core of EnViTSA resides in its ensemble of pre-trained Vision Transformers, harnessing the unique strengths of the Vision Transformer architecture. This ensemble approach not only reduces inductive biases but also effectively mitigates overfitting. In this study, we evaluate the EnViTSA method on three benchmark datasets: ESC-10, ESC-50, and UrbanSound8K. The experimental results underscore the efficacy of our approach, achieving impressive accuracy scores of 93.50%, 85.85%, and 83.20% on ESC-10, ESC-50, and UrbanSound8K, respectively. EnViTSA represents a substantial advancement in AEC, demonstrating the potential of Vision Transformers and SpecAugment in the acoustic domain. [ABSTRACT FROM AUTHOR]

In this paper, research was conducted on anomaly detection of wheel flats. In the railway sector, conducting tests with actual railway vehicles is challenging due to safety concerns for passengers and maintenance issues as it is a public industry. Therefore, dynamics software was utilized. Next, STFT (short-time Fourier transform) was performed to create spectrogram images. In the case of railway vehicles, control, monitoring, and communication are performed through TCMS, but complex analysis and data processing are difficult because there are no devices such as GPUs. Furthermore, there are memory limitations. Therefore, in this paper, the relatively lightweight models LeNet-5, ResNet-20, and MobileNet-V3 were selected for deep learning experiments. At this time, the LeNet-5 and MobileNet-V3 models were modified from the basic architecture. Since railway vehicles are given preventive maintenance, it is difficult to obtain fault data. Therefore, semi-supervised learning was also performed. At this time, the Deep One Class Classification paper was referenced. The evaluation results indicated that the modified LeNet-5 and MobileNet-V3 models achieved approximately 97% and 96% accuracy, respectively. At this point, the LeNet-5 model showed a training time of 12 min faster than the MobileNet-V3 model. In addition, the semi-supervised learning results showed a significant outcome of approximately 94% accuracy when considering the railway maintenance environment. In conclusion, considering the railway vehicle maintenance environment and device specifications, it was inferred that the relatively simple and lightweight LeNet-5 model can be effectively utilized while using small images. [ABSTRACT FROM AUTHOR]

We propose a method based on Synchrosqueezing Transform (SST) for vibration event analysis and identification in Phase Sensitive Optical Time-Domain Reflectometry (Φ-OTDR) systems. SST has high time-frequency resolution and phase information, which can distinguish and enhance different vibration events. We use six tap events with different intensities and six other events as experimental data and test the effect of attenuation. We use Visual Geometry Group (VGG), Vision Transformer (ViT), and Residual Network (ResNet) as deep classifiers for the SST transformed data. The results show that our method outperforms the methods based on Continuous Wavelet Transform (CWT) and Short-Time Fourier Transform (STFT), while ResNet is the best classifier. Our method can achieve high recognition rate under different signal strengths, event types, and attenuation levels, which shows its value for Φ-OTDR system. [ABSTRACT FROM AUTHOR]

Source counting is the key procedure of autonomous detection for underwater unmanned platforms. A source counting method with local-confidence-level-enhanced density clustering using a single acoustic vector sensor (AVS) is proposed in this paper. The short-time Fourier transforms (STFT) of the sound pressure and vibration velocity measured by the AVS are first calculated, and a data set is established with the direction of arrivals (DOAs) estimated from all of the time–frequency points. Then, the density clustering algorithm is used to classify the DOAs in the data set, with which the number of the clusters and the cluster centers are obtained as the source number and the DOA estimations, respectively. In particular, the local confidence level is adopted to weigh the density of each DOA data point to highlight samples with the dominant sources and downplay those without, so that the differences in densities for the cluster centers and sidelobes are increased. Therefore, the performance of the density clustering algorithm is improved, leading to an improved source counting accuracy. Experimental results reveal that the enhanced source counting method achieves a better source counting performance than that of basic density clustering. [ABSTRACT FROM AUTHOR]

The ultrasonic guided wave technique is extensively used for nondestructive structural testing, and one of the key steps is to extract a single mode with certain purity from multi-order mixed modes. In this paper, the propagation of ultrasonic guided waves in the cylindrical rod is simulated first; the appropriate broadband excitation signal is selected to excite the multi-order modes in a specific frequency range; and the time–space signal containing multi-order modes is converted to the frequency-wavenumber domain signal by two-dimensional Fourier transform. In the frequency-wavenumber domain, the frequency-wavenumber ridge is extracted from the multi-mode frequency-wavenumber domain based on the dynamic programming method, and then the time-domain signal corresponding to a single mode can be reconstructed. By comparing the excited multi-order mode and the separated single mode with the theoretical results, it is observed that the two results are consistent. Thus, the employed mode-excitation method can accurately excite the multi-order modes in rod structures. Furthermore, the proposed method enables the separation of a single-mode wave with high purity, providing a foundation for future utilization of isolated modes. [ABSTRACT FROM AUTHOR]

A hybrid deep learning approach was designed that combines deep learning with enhanced short-time Fourier transform (STFT) spectrograms and continuous wavelet transform (CWT) scalograms for pipeline leak detection. Such detection plays a crucial role in ensuring the safety and integrity of fluid transportation systems. The proposed model leverages the power of STFT and CWT to enhance detection capabilities. The pipeline's acoustic emission signals during normal and leak operating conditions undergo transformation using STFT and CWT, creating scalograms representing energy variations across time–frequency scales. To improve the signal quality and eliminate noise, Sobel and wavelet denoising filters are applied to the scalograms. These filtered scalograms are then fed into convolutional neural networks, extracting informative features that harness the distinct characteristics captured by both STFT and CWT. For enhanced computational efficiency and discriminatory power, principal component analysis is employed to reduce the feature space dimensionality. Subsequently, pipeline leaks are accurately detected and classified by categorizing the reduced dimensional features using t-distributed stochastic neighbor embedding and artificial neural networks. The hybrid approach achieves high accuracy and reliability in leak detection, demonstrating its effectiveness in capturing both spectral and temporal details. This research significantly contributes to pipeline monitoring and maintenance and offers a promising solution for real-time leak detection in diverse industrial applications. [ABSTRACT FROM AUTHOR]

A microwave absolute distance measurement method with ten-micron-level accuracy and meter-level range based on frequency domain interferometry is proposed and experimentally demonstrated for the first time. Theoretical analysis indicates that an interference phenomenon occurs instantaneously in the frequency domain when combining two homologous broad-spectrum microwave beams with different paths, and the absolute value of the distance difference between the two paths is only inversely proportional to the period of frequency domain interference fringes. The proof-of-principle experiments were performed to prove that the proposed method can achieve absolute distance measurement in the X-band with standard deviations of 15 μm, 17 μm, and 26 μm and within ranges of 1.69 m, 2.69 m, and 3.75 m. Additionally, a displacement resolution of 100 microns was realized. The multi-target recognition performance was also verified in principle. Furthermore, at the expense of a slight decrease in ranging accuracy, a fast distance measurement with the single measurement time of 20 μs was achieved by using a digitizer combined with a Fourier transform analyzer. Compared with the current microwave precision ranging technologies, the proposed method not only has the advantages of high precision, large range, and rapid measurement capability, but the required components are also easily obtainable commercial devices. The proposed method also has better complex engineering applicability, because the ten-micron-level ranging accuracy is achievable only by using a simple Fourier transform without any phase estimation algorithm, which greatly reduces the requirement for signal-to-noise ratio. [ABSTRACT FROM AUTHOR]

In low-voltage distribution systems, the load types are complex, so traditional detection methods cannot effectively identify series arc faults. To address this problem, this paper proposes an arc fault detection method based on multimodal feature fusion. Firstly, the different mode features of the current signal are extracted by mathematical statistics, Fourier transform, wavelet packet transform, and continuous wavelet transform. The different modal features include one-dimensional features, such as time-domain features, frequency-domain features, and wavelet packet energy features, and two-dimensional features of time-spectrum images. Secondly, the extracted features are preprocessed and prioritized for importance based on different machine learning algorithms to improve the feature data quality. The features of higher importance are input into an arc fault detection model. Finally, an arc fault detection model is constructed based on a one-dimensional convolutional network and a deep residual shrinkage network to achieve high accuracy. The proposed detection method has higher detection accuracy and better performance compared with the arc fault detection method based on single-mode features. [ABSTRACT FROM AUTHOR]

Near-infrared (NIR) photodetectors (PDs) have attracted much attention for use in noninvasive medical diagnosis and treatments. In particular, self-filtered NIR PDs are in high demand for a wide range of biomedical applications due to their ability for wavelength discrimination. In this work, we designed and then fabricated a Si micro-hole array/Graphene (Si MHA/Gr) van der Waals (vdW) Schottky NIR photodiode using a PbS quantum dot (QD) coating. The device exhibited a unique self-filtered NIR response with a responsivity of 0.7 A/W at −1 V and a response speed of 61 μs, which is higher than that seen without PbS QD coating and even in most previous Si/Gr Schottky photodiodes. The light trapping of the Si MHA and the PbS QD coating could be attributed to the high responsivity of the vdW photodiode. Furthermore, the presented NIR photodiode could also be integrated in photoplethysmography (PPG) for real-time heart rate (HR) monitoring. The extracted HR was in good accord with the values measured with the patient monitor—determined by analyzing the Fourier transform of the stable and reliable fingertip PPG waveform—suggesting its potential for practical applications. [ABSTRACT FROM AUTHOR]

Roll eccentricity disturbance is a high-frequency periodic change signal caused by the irregular shape of the roll and roll bearing, which is difficult to identify and affects the periodic deviation of the exit thickness of the strip. To achieve rapid identification of the source and a mathematical model of roll eccentricity signals, a sparse Fourier transform (SFT) and regional DFT method for roll eccentricity signal recognition and detection was proposed. This method utilizes SFT to calculate the signal FFT more quickly based on the sparsity of the signal frequency domain. Under the premise of knowing the roll diameter, the signal frequency spectrum is identified online, the amplitude and phase are identified through regional DFT, and the eccentricity disturbance is compensated on site. The simulation results show that this method can accurately identify the source of roll disturbance, quickly update and replace the problematic rolls, and improve the online recognition efficiency by more than 3000 times. This method has good results in online detection and recognition of roll eccentricity signals, greatly improving engineering application efficiency, and ultimately achieving the goal of improving the accuracy of strip outlet thickness. [ABSTRACT FROM AUTHOR]

Micro electro-mechanical systems (MEMS) combining sensing and microfluidics functionalities, as are common in Lab-on-Chip (LoC) devices, are increasingly based on polymers. Benefits of polymers include tunable material properties, the possibility of surface functionalization, compatibility with many micro and nano patterning techniques, and optical transparency. Often, additional materials, such as metals, ceramics, or silicon, are needed for functional or auxiliary purposes, e.g., as electrodes. Hybrid patterning and integration of material composites require an increasing range of fabrication approaches, which must often be newly developed or at least adapted and optimized. Here, a microfabrication process concept is developed that allows one to implement attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR) and electrochemistry on an LoC device. It is designed to spatially resolve chemical sensitivity and selectivity, which are instrumental for the detection of chemical distributions, e.g., during on-flow chemical and biological reaction chemistry. The processing sequence involves (i) direct-write and soft-contact UV lithography in SUEX dry resist and replication in polydimethylsiloxane (PDMS) elastomers as the fluidic structure; (ii) surface functionalization of PDMS with oxygen plasma, 3-aminopropyl-triethoxysilane (APTES), and a UV-curable glue (NOA 73) for bonding the fluidic structure to the substrate; (iii) double-sided patterning of silicon nitride-coated silicon wafers serving as the ATR-FTIR-active internal reflection element (IRE) on one side and the electrode-covered substrate for microfluidics on the back side with lift-off and sputter-based patterning of gold electrodes; and (iv) a custom-designed active vacuum positioning and alignment setup. Fluidic channels of 100 μm height and 600 μm width in 5 mm thick PDMS were fabricated on 2" and 4" demonstrators. Electrochemistry on-chip functionality was demonstrated by cyclic voltammetry (CV) of redox reactions involving iron cyanides in different oxidation states. Further, ATR-FTIR measurements of laminar co-flows of H2O and D2O demonstrated the chemical mapping capabilities of the modular fabrication concept of the LoC devices. [ABSTRACT FROM AUTHOR]

Multiple-input-multiple-output synthetic aperture radar (MIMO-SAR) is being studied and applied in more and more scenarios. However, there is still a certain distance away from real-time imaging using advanced algorithms. The traditional backpropagation algorithm (BPA) multi-accumulation integration is unsuitable for dealing with large-size scanning data, and the wavenumber domain algorithm requires the array to satisfy Nyquist sampling law in azimuth to avoid aliasing in imaging reconstruction. Based on these issues, a novel 3D imaging method is proposed for MIMO-SAR. An appropriate transformation and inverse Fourier transform (FT) is carried out for the frequency domain; thus, accumulation in the wavenumber domain is not required, which is easy to implement. The computational complexity of the algorithm is much lower than BPA and has better generalizability than the wavenumber domain algorithm. Coherence factor (CF) is also introduced to achieve sidelobe suppression. Proof-of-principle experiments were also carried out in the 92.5 GHz band based on the MIMO-SAR prototype system. Both simulation and experimental results of different distributed targets show good performance of imaging and do not lose the quality of image reconstruction. [ABSTRACT FROM AUTHOR]

Deaf and hearing-impaired people always face communication barriers. Non-invasive surface electromyography (sEMG) sensor-based sign language recognition (SLR) technology can help them to better integrate into social life. Since the traditional tandem convolutional neural network (CNN) structure used in most CNN-based studies inadequately captures the features of the input data, we propose a novel inception architecture with a residual module and dilated convolution (IRDC-net) to enlarge the receptive fields and enrich the feature maps, applying it to SLR tasks for the first time. This work first transformed the time domain signal into a time–frequency domain using discrete Fourier transformation. Second, an IRDC-net was constructed to recognize ten Chinese sign language signs. Third, the tandem CNN networks VGG-net and ResNet-18 were compared with our proposed parallel structure network, IRDC-net. Finally, the public dataset Ninapro DB1 was utilized to verify the generalization performance of the IRDC-net. The results showed that after transforming the time domain sEMG signal into the time–frequency domain, the classification accuracy (acc) increased from 84.29% to 91.70% when using the IRDC-net on our sign language dataset. Furthermore, for the time–frequency information of the public dataset Ninapro DB1, the classification accuracy reached 89.82%; this value is higher than that achieved in other recent studies. As such, our findings contribute to research into SLR tasks and to improving deaf and hearing-impaired people's daily lives. [ABSTRACT FROM AUTHOR]

Tool wear condition monitoring is an important component of mechanical processing automation, and accurately identifying the wear status of tools can improve processing quality and production efficiency. This paper studied a new deep learning model, to identify the wear status of tools. The force signal was transformed into a two-dimensional image using continuous wavelet transform (CWT), short-time Fourier transform (STFT), and Gramian angular summation field (GASF) methods. The generated images were then fed into the proposed convolutional neural network (CNN) model for further analysis. The calculation results show that the accuracy of tool wear state recognition proposed in this paper was above 90%, which was higher than the accuracy of AlexNet, ResNet, and other models. The accuracy of the images generated using the CWT method and identified with the CNN model was the highest, which is attributed to the fact that the CWT method can extract local features of an image and is less affected by noise. Comparing the precision and recall values of the model, it was verified that the image obtained by the CWT method had the highest accuracy in identifying tool wear state. These results demonstrate the potential advantages of using a force signal transformed into a two-dimensional image for tool wear state recognition and of applying CNN models in this area. They also indicate the wide application prospects of this method in industrial production. [ABSTRACT FROM AUTHOR]

A frequency-modulated continuous-wave radar for short-range target imaging, assembling a transceiver, a PLL, an SP4T switch, and a serial patch antenna array, was realized. A new algorithm based on a double Fourier transform (2D-FT) was developed and compared with the delay and sum (DAS) and multiple signal classification (MUSIC) algorithms proposed in the literature for target detection. The three reconstruction algorithms were applied to simulated canonical cases evidencing radar resolutions close to the theoretical ones. The proposed 2D-FT algorithm exhibits an angle of view greater than 25° and is five times faster than DAS and 20 times faster than the MUSIC one. The realized radar shows a range resolution of 55 cm and an angular resolution of 14° and is able to correctly identify the positions of single and multiple targets in realistic scenarios, with errors lower than 20 cm. [ABSTRACT FROM AUTHOR]

The detection of anomalies in multivariate time-series data is becoming increasingly important in the automated and continuous monitoring of complex systems and devices due to the rapid increase in data volume and dimension. To address this challenge, we present a multivariate time-series anomaly detection model based on a dual-channel feature extraction module. The module focuses on the spatial and time features of the multivariate data using spatial short-time Fourier transform (STFT) and a graph attention network, respectively. The two features are then fused to significantly improve the model's anomaly detection performance. In addition, the model incorporates the Huber loss function to enhance its robustness. A comparative study of the proposed model with existing state-of-the-art ones was presented to prove the effectiveness of the proposed model on three public datasets. Furthermore, by using in shield tunneling applications, we verify the effectiveness and practicality of the model. [ABSTRACT FROM AUTHOR]

Frequency estimation plays a critical role in vital sign monitoring. Methods based on Fourier transform and eigen-analysis are commonly adopted techniques for frequency estimation. Because of the nonstationary and time-varying characteristics of physiological processes, time-frequency analysis (TFA) is a feasible way to perform biomedical signal analysis. Among miscellaneous approaches, Hilbert–Huang transform (HHT) has been demonstrated to be a potential tool in biomedical applications. However, the problems of mode mixing, unnecessary redundant decomposition and boundary effect are the common deficits that occur during the procedure of empirical mode decomposition (EMD) or ensemble empirical mode decomposition (EEMD). The Gaussian average filtering decomposition (GAFD) technique has been shown to be appropriate in several biomedical scenarios and can be an alternative to EMD and EEMD. This research proposes the combination of GAFD and Hilbert transform that is termed the Hilbert–Gauss transform (HGT) to overcome the conventional drawbacks of HHT in TFA and frequency estimation. This new method is verified to be effective for the estimation of respiratory rate (RR) in finger photoplethysmography (PPG), wrist PPG and seismocardiogram (SCG). Compared with the ground truth values, the estimated RRs are evaluated to be of excellent reliability by intraclass correlation coefficient (ICC) and to be of high agreement by Bland–Altman analysis. [ABSTRACT FROM AUTHOR]

To estimate the direction of arrival (DOA) of a linear frequency modulation (LFM) signal in a low signal-to-noise ratio (SNR) hydroacoustic environment by a small aperture array, a novel deconvolved beamforming method based on fractional Fourier domain delay-and-sum beamforming (FrFB) was proposed. Fractional Fourier transform (FrFT) was used to convert the received signal into the fractional Fourier domain, and delay-and-sum beamforming was subsequently performed. Noise resistance was acquired by focusing the energy of the LFM signal distributed in the time–frequency domain. Then, according to the convolution structure of the FrFB complex output, the influence of the fractional Fourier domain complex beam pattern was removed by deconvolution, and the target spatial distribution was restored. Therefore, an improved spatial resolution of DOA estimation was obtained without increasing the array aperture. The simulation and experimental results show that, with a small aperture array at low SNR, the proposed method possesses higher spatial resolution than FrFB and frequency-domain deconvolved conventional beamforming. [ABSTRACT FROM AUTHOR]

A transformer's acoustic signal contains rich information. The acoustic signal can be divided into a transient acoustic signal and a steady-state acoustic signal under different operating conditions. In this paper, the vibration mechanism is analyzed, and the acoustic feature is mined based on the transformer end pad falling defect to realize defect identification. Firstly, a quality–spring–damping model is established to analyze the vibration modes and development patterns of the defect. Secondly, short-time Fourier transform is applied to the voiceprint signals, and the time–frequency spectrum is compressed and perceived using Mel filter banks. Thirdly, the time-series spectrum entropy feature extraction algorithm is introduced into the stability calculation, and the algorithm is verified by comparing it with simulated experimental samples. Finally, stability calculations are performed on the voiceprint signal data collected from 162 transformers operating in the field, and the stability distribution is statistically analyzed. The time-series spectrum entropy stability warning threshold is given, and the application value of the threshold is demonstrated by comparing it with actual fault cases. [ABSTRACT FROM AUTHOR]

Fault diagnosis is important in rotor systems because severe damage can occur during the operation of systems under harsh conditions. The advancements in machine learning and deep learning have led to enhanced performance of classification. Two important elements of fault diagnosis using machine learning are data preprocessing and model structure. Multi–class classification is used to classify faults into different single types, whereas multi–label classification classifies faults into compound types. It is valuable to focus on the capability of detecting compound faults because multiple faults can exist simultaneously. Diagnosis of untrained compound faults is also a merit. In this study, input data were first preprocessed with short–time Fourier transform. Then, a model was built for classification of the state of the system based on multi–output classification. Finally, the proposed model was evaluated based on its performance and robustness for classification of compound faults. This study proposes an effective model based on multi–output classification, which can be trained using only single fault data for the classification of compound faults and confirms the robustness of the model to changes in unbalance. [ABSTRACT FROM AUTHOR]

Determining the presence and severity of knee osteoarthritis (OA) is a valuable application of inertial measurement units (IMUs) in the remote monitoring of patients. This study aimed to employ the Fourier representation of IMU signals to differentiate between individuals with and without knee OA. We included 27 patients with unilateral knee osteoarthritis (15 females) and 18 healthy controls (11 females). Gait acceleration signals were recorded during overground walking. We obtained the frequency features of the signals using the Fourier transform. The logistic LASSO regression was employed on the frequency domain features as well as the participant's age, sex, and BMI to distinguish between the acceleration data from individuals with and without knee OA. The model's accuracy was estimated by 10-fold cross-validation. The frequency contents of the signals were different between the two groups. The average accuracy of the classification model using the frequency features was 0.91 ± 0.01. The distribution of the selected features in the final model differed between patients with different severity of knee OA. In this study, we demonstrated that using logistic LASSO regression on the Fourier representation of acceleration signals can accurately determine the presence of knee OA. [ABSTRACT FROM AUTHOR]

This article analyses the possibility of using the Analytic Wavelet Transform (AWT) and the Convolutional Neural Network (CNN) for the purpose of recognizing the intrapulse modulation of radar signals. Firstly, the possibilities of using AWT by the algorithms of automatic signal recognition are discussed. Then, the research focuses on the influence of the parameters of the generalized Morse wavelet on the classification accuracy. The paper's novelty is also related to the use of the generalized Morse wavelet (GMW) as a superfamily of analytical wavelets with a Convolutional Neural Network (CNN) as classifier applied for intrapulse recognition purposes. GWT is used to obtain time–frequency images (TFI), and SqueezeNet was chosen as the CNN classifier. The article takes into account selected types of intrapulse modulation, namely linear frequency modulation (LFM) and the following types of phase-coded waveform (PCW): Frank, Barker, P1, P2, and Px. The authors also consider the possibility of using other time–frequency transformations such as Short-Time Fourier Transform(STFT) or Wigner–Ville Distribution (WVD). Finally, authors present the results of the simulation tests carried out in the Matlab environment, taking into account the signal-to-noise ratio (SNR) in the range from −6 to 0 dB. [ABSTRACT FROM AUTHOR]

In this work, we present a new nonlinear joint transform correlator (JTC) architecture in the Fourier domain (FD) for the encryption and decryption of two simultaneous images. The main features of the proposed system are its increased level of security, the obtention of a single real-valued encrypted signal that contains the ciphered information of the two primary images and, additionally, a high image quality for the two final decrypted signals. The two images to be encrypted can be either related to each other, or independent signals. The encryption system is based on the double random phase encoding (DRPE), which is implemented by using a nonlinear JTC in the FD. The input plane of the JTC has four non-overlapping data distributions placed side-by-side with no blank spaces between them. The four data distributions are phase-only functions defined by the two images to encrypt and four random phase masks (RPMs). The joint power spectrum (JPS) is produced by the intensity of the Fourier transform (FT) of the input plane of the JTC. One of the main novelties of the proposal consists of the determination of the appropriate two nonlinear operations that modify the JPS distribution with a twofold purpose: to obtain a single real-valued encrypted image with a high level of security and to improve the quality of the decrypted images. The security keys of the encryption system are represented by the four RPMs, which are all necessary for a satisfactory decryption. The decryption system is implemented using a 4f-processor where the encrypted image and the security keys given by the four RPMs are introduced in the proper plane of the processor. The double image encryption system based on a nonlinear JTC in the FD increases the security of the system because there is a larger key space, and we can simultaneously validate two independent information signals (original images to encrypt) in comparison to previous similar proposals. The feasibility and performance of the proposed double image encryption and decryption system based on a nonlinear JTC are validated through computational simulations. Finally, we additionally comment on the proposed security system resistance against different attacks based on brute force, plaintext and deep learning. [ABSTRACT FROM AUTHOR]

With the development of wireless technology, signals propagating in space are easy to mix, so blind detection of communication signals has become a very practical and challenging problem. In this paper, we propose a blind detection method for broadband signals based on a weighted bi-directional feature pyramid network (BiFPN). The method can quickly perform detection and automatic modulation identification (AMC) on time-domain aliased signals in broadband data. Firstly, the method performs a time-frequency analysis on the received signals and extracts the normalized time-frequency images and the corresponding labels by short-time Fourier transform (STFT). Secondly, we build a target detection model based on YOLOv5 for time-domain mixed signals in broadband data and learn the features of the time-frequency distribution image dataset of broadband signals, which achieves the purpose of training the model. The main improvements of the algorithm are as follows: (1) a weighted bi-directional feature pyramid network is used to achieve a simple and fast multi-scale feature fusion approach to improve the detection probability; (2) the Efficient-Intersection over Union (EIOU) loss function is introduced to achieve high accuracy signal detection in a low Signal-Noise Ratio (SNR) environment. Finally, the time-frequency images are detected by an improved deep network model to complete the blind detection of time-domain mixed signals. The simulation results show that the method can effectively detect the continuous and burst signals in the broadband communication signal data and identify their modulation types. [ABSTRACT FROM AUTHOR]

The direction of arrival (DOA) and number of sound sources is usually estimated by short-time Fourier transform and the conjugate cross-spectrum. However, the ability of a single AVS to distinguish between multiple sources will decrease as the number of sources increases. To solve this problem, this paper presents a multimodal fusion method based on a single acoustic vector sensor (AVS). First, the output of the AVS is decomposed into multiple modes by intrinsic time-scale decomposition (ITD). The number of sources in each mode decreases after decomposition. Then, the DOAs and source number in each mode are estimated by density peak clustering (DPC). Finally, the density-based spatial clustering of applications with the noise (DBSCAN) algorithm is employed to obtain the final source counting results from the DOAs of all modes. Experiments showed that the multimodal fusion method could significantly improve the ability of a single AVS to distinguish multiple sources when compared to methods without multimodal fusion. [ABSTRACT FROM AUTHOR]

Optical encoders are widely used in accurate displacement measurement and motion-control technologies. Based on different measurement methods, optical encoders can be divided into absolute and incremental optical encoders. Absolute linear encoders are commonly used in advanced computer numerical control (CNC) machines. The subdivision error within one signal period (SDE) of the absolute linear encoder is vital to the positioning accuracy and low velocity control of CNC machines. In our paper, we study the working principle of the absolute linear encoder. We proposed two methods for reducing the SDE of the absolute linear encoder, a single-field scanning method based on the shutter-shaped Moiré fringe, as well as a method for suppressing harmonics through a phase shift of index grating. We established a SDE measuring device to determine the absolute linear encoder's SDE, which we measured using a constant-speed approach. With our proposed methods, the SDE was reduced from ±0.218 μm to ±0.135 μm, which is a decrease of 38.07%. Our fast Fourier transformation (FFT) analysis of the collected Moiré fringe signals demonstrated that the third-, fifth-, and seventh-order harmonics were effectively suppressed. [ABSTRACT FROM AUTHOR]

Personal identification based on radar gait measurement is an important application of biometric technology because it enables remote and continuous identification of people, irrespective of the lighting conditions and subjects' outfits. This study explores an effective time-velocity distribution and its relevant parameters for Doppler-radar-based personal gait identification using deep learning. Most conventional studies on radar-based gait identification used a short-time Fourier transform (STFT), which is a general method to obtain time-velocity distribution for motion recognition using Doppler radar. However, the length of the window function that controls the time and velocity resolutions of the time-velocity image was empirically selected, and several other methods for calculating high-resolution time-velocity distributions were not considered. In this study, we compared four types of representative time-velocity distributions calculated from the Doppler-radar-received signals: STFT, wavelet transform, Wigner–Ville distribution, and smoothed pseudo-Wigner–Ville distribution. In addition, the identification accuracies of various parameter settings were also investigated. We observed that the optimally tuned STFT outperformed other high-resolution distributions, and a short length of the window function in the STFT process led to a reasonable accuracy; the best identification accuracy was 99% for the identification of twenty-five test subjects. These results indicate that STFT is the optimal time-velocity distribution for gait-based personal identification using the Doppler radar, although the time and velocity resolutions of the other methods were better than those of the STFT. [ABSTRACT FROM AUTHOR]

The modulation transfer function (MTF) is an important parameter for performance evaluation of optical imaging systems in photogrammetry and remote sensing; the slanted-edge method is one of the main methods for measuring MTF. To solve the problem of inaccurate edge detection by traditional methods under the conditions of noise and blur, this paper proposes a new method of MTF measurement with a slanted-edge method based on improved Zernike moments, which firstly introduces the Otsu algorithm to automatically determine the Zernike moment threshold for sub-pixel edge detection to precisely locate the edge points, then obtains LSF through edge point projection, ESF sampling point acquisition, smoothing, fitting, taking ESF curve differential and Gaussian fitting, and finally, accurately obtaining MTF by LSF Fourier transform and modulo normalization. Based on simulation experiments and outdoor target experiments, the reliability of the proposed algorithm is verified by the deviations of slanted-edge angle and MTF measurement, and the tolerance degree of edge detection to noise and ambiguity are analyzed. The results show that compared with ISO 12233, OMNI-sine method, Hough transform method and LSD method, this algorithm has the highest edge detection accuracy, the maximum tolerance of noise and ambiguity, and also improves the accuracy of MTF measurement. [ABSTRACT FROM AUTHOR]

In push-broom hyperspectral imaging systems, the sensor rotation to the optical plane leads to linear spatial misregistration (LSM) in hyperspectral images (HSIs). To compensate for hardware defects through software, this paper develops four methods to detect LSM in HSIs. Different from traditional methods for grayscale images, the method of fitting the sum of abundance (FSAM) and the method of searching for equal abundance (SEAM) are achieved by hyperspectral unmixing for a selected rectangular transition areas containing an edge, which makes good use of spatial and spectral information. The method based on line detection for band-interleaved-by-line (BIL) images (LDBM) and the method based on the Fourier transform of BIL images (FTBM) aim to characterize the slope of line structure in BIL images and get rid of the dependence on scene and wavelength. A full strategy is detailed from aspects of data selection, LSM detection, and image correction. The full spectrum airborne hyperspectral imager (FAHI) is China's new generation push-broom scanner. The HSIs obtained by FAHI are tested and analyzed. Experiments on simulation data compare the four proposed methods with traditional methods and prove that FSAM outperforms other methods in terms of accuracy and stability. In experiments on real data, the application of the full strategy on FAHI verifies its effectiveness. This work not only provides reference for other push-broom imagers with similar problems, but also helps to reduce the requirement for hardware calibration. [ABSTRACT FROM AUTHOR]

From the point of view of measurement, footstep sounds represent a simple, wearable and inexpensive sensing opportunity to assess running biomechanical parameters. Therefore, the aim of this study was to investigate whether the sounds of footsteps can be used to predict the vertical ground reaction force profiles during running. Thirty-seven recreational runners performed overground running, and their sounds of footsteps were recorded from four microphones, while the vertical ground reaction force was recorded using a force plate. We generated nine different combinations of microphone data, ranging from individual recordings up to all four microphones combined. We trained machine learning models using these microphone combinations and predicted the ground reaction force profiles by a leave-one-out approach on the subject level. There were no significant differences in the prediction accuracy between the different microphone combinations (p < 0.05). Moreover, the machine learning model was able to predict the ground reaction force profiles with a mean Pearson correlation coefficient of 0.99 (range 0.79–0.999), mean relative root-mean-square error of 9.96% (range 2–23%) and mean accuracy to define rearfoot or forefoot strike of 77%. Our results demonstrate the feasibility of using the sounds of footsteps in combination with machine learning algorithms based on Fourier transforms to predict the ground reaction force curves. The results are encouraging in terms of the opportunity to create wearable technology to assess the ground reaction force profiles for runners in the interests of injury prevention and performance optimization. [ABSTRACT FROM AUTHOR]

The non-invasive electrocardiogram (ECG) signals are useful in heart condition assessment and are found helpful in diagnosing cardiac diseases. However, traditional ways, i.e., a medical consultation required effort, knowledge, and time to interpret the ECG signals due to the large amount of data and complexity. Neural networks have been shown to be efficient recently in interpreting the biomedical signals including ECG and EEG. The novelty of the proposed work is using spectrograms instead of raw signals. Spectrograms could be easily reduced by eliminating frequencies with no ECG information. Moreover, spectrogram calculation is time-efficient through short-time Fourier transformation (STFT) which allowed to present reduced data with well-distinguishable form to convolutional neural network (CNN). The data reduction was performed through frequency filtration by taking a specific cutoff value. These steps makes architecture of the CNN model simple which showed high accuracy. The proposed approach reduced memory usage and computational power through not using complex CNN models. A large publicly available PTB-XL dataset was utilized, and two datasets were prepared, i.e., spectrograms and raw signals for binary classification. The highest accuracy of 99.06% was achieved by the proposed approach, which reflects spectrograms are better than the raw signals for ECG classification. Further, up- and down-sampling of the signals were also performed at various sampling rates and accuracies were attained. [ABSTRACT FROM AUTHOR]

As a promising alternative to conventional contact sensors, vision-based technologies for a structural dynamic response measurement and health monitoring have attracted much attention from the research community. Among these technologies, Eulerian video magnification has a unique capability of analyzing modal responses and visualizing modal shapes. To reduce the noise interference and improve the quality and stability of the modal shape visualization, this study proposes a hybrid motion magnification framework that combines linear and phase-based motion processing. Based on the assumption that temporal variations can represent spatial motions, the linear motion processing extracts and manipulates the temporal intensity variations related to modal responses through matrix decomposition and underdetermined blind source separation (BSS) techniques. Meanwhile, the theory of Fourier transform profilometry (FTP) is utilized to reduce spatial high-frequency noise. As all spatial motions in a video are linearly controllable, the subsequent phase-based motion processing highlights the motions and visualizes the modal shapes with a higher quality. The proposed method is validated by two laboratory experiments and a field test on a large-scale truss bridge. The quantitative evaluation results with high-speed cameras demonstrate that the hybrid method performs better than the single-step phase-based motion magnification method in visualizing sound-induced subtle motions. In the field test, the vibration characteristics of the truss bridge when a train is driving across the bridge are studied with a commercial camera over 400 m away from the bridge. Moreover, four full-field modal shapes of the bridge are successfully observed. [ABSTRACT FROM AUTHOR]

Landslide susceptibility mapping (LSM) is of great significance for the identification and prevention of geological hazards. LSM is based on convolutional neural networks (CNNs); CNNs use fixed convolutional kernels, focus more on local information and do not retain spatial information. This is a property of the CNN itself, resulting in low accuracy of LSM. Based on the above problems, we use Vision Transformer (ViT) and its derivative model Swin Transformer (Swin) to conduct LSM for the selected study area. Machine learning and a CNN model are used for comparison. Fourier transform amplitude, feature similarity and other indicators were used to compare and analyze the difference in the results. The results show that the Swin model has the best accuracy, F1-score and AUC. The results of LSM are combined with landslide points, faults and other data analysis; the ViT model results are the most consistent with the actual situation, showing the strongest generalization ability. In this paper, we believe that the advantages of ViT and its derived models in global feature extraction ensure that ViT is more accurate than CNN and machine learning in predicting landslide probability in the study area. [ABSTRACT FROM AUTHOR]

The induction motor plays a vital role in industrial drive systems due to its robustness and easy maintenance but at the same time, it suffers electrical faults, mainly rotor faults such as broken rotor bars. Early shortcoming identification is needed to lessen support expenses and hinder high costs by using failure detection frameworks that give features extraction and pattern grouping of the issue to distinguish the failure in an induction motor using classification models. In this paper, the open-source dataset of the rotor with the broken bars in a three-phase induction motor available on the IEEE data port is used for fault classification. The study aims at fault identification under various loading conditions on the rotor of an induction motor by performing time, frequency, and time-frequency domain feature extraction. The extracted features are provided to the models to classify between the healthy and faulty rotors. The extracted features from the time and frequency domain give an accuracy of up to 87.52% and 88.58%, respectively, using the Random-Forest (RF) model. Whereas, in time-frequency, the Short Time Fourier Transform (STFT) based spectrograms provide reasonably high accuracy, around 97.67%, using a Convolutional Neural Network (CNN) based fine-tuned transfer learning framework for diagnosing induction motor rotor bar severity under various loading conditions. [ABSTRACT FROM AUTHOR]

Multiphase machines have recently been promoted as a viable alternative to traditional three-phase machines. Most experts are looking for strategies to estimate the rotation speed of such complex systems, since speed data are required for high-performance control purposes. Traditionally, electromechanical sensors were used to detect the rotor speed of electric motors. These devices are extremely accurate, but they are also delicate and costly to deploy. New speed estimating algorithms must be created for these situations. This paper looks at how to estimate rotor speed in symmetrical six-phase induction motors (IMs) using a novel strategy for rotor speed estimation based on the Short Time Fourier Transform (STFT) method. The technique is based on tracking the frequencies of the rotor slot harmonics (RSH) seen in most squirrel-cage IM stator currents, thus assuring a broad range of applications. To monitor the RSH, the STFT employs a sliding window to perform the discrete Fourier transform technique, making it more suitable for online use with noisy and nonstationary signals. Experimental tests demonstrate the effectiveness of the suggested approach. [ABSTRACT FROM AUTHOR]

In environment sound classification, log Mel band energies (MBEs) are considered as the most successful and commonly used features for classification. The underlying algorithm, fast Fourier transform (FFT), is valid under certain restrictions. In this study, we address these limitations of Fourier transform and propose a new method to extract log Mel band energies using amplitude modulation and frequency modulation. We present a comparative study between traditionally used log Mel band energy features extracted by Fourier transform and log Mel band energy features extracted by our new approach. This approach is based on extracting log Mel band energies from estimation of instantaneous frequency (IF) and instantaneous amplitude (IA), which are used to construct a spectrogram. The estimation of IA and IF is made by associating empirical mode decomposition (EMD) with the Teager–Kaiser energy operator (TKEO) and the discrete energy separation algorithm. Later, Mel filter bank is applied to the estimated spectrogram to generate EMD-TKEO-based MBEs, or simply, EMD-MBEs. In addition, we employ the EMD method to remove signal trends from the original signal and generate another type of MBE, called S-MBEs, using FFT and a Mel filter bank. Four different datasets were utilised and convolutional neural networks (CNN) were trained using features extracted from Fourier transform-based MBEs (FFT-MBEs), EMD-MBEs, and S-MBEs. In addition, CNNs were trained with an aggregation of all three feature extraction techniques and a combination of FFT-MBEs and EMD-MBEs. Individually, FFT-MBEs achieved higher accuracy compared to EMD-MBEs and S-MBEs. In general, the system trained with the combination of all three features performed slightly better compared to the system trained with the three features separately. [ABSTRACT FROM AUTHOR]

In modern industrial production, the prediction ability of remaining useful life of bearings directly affects the safety and stability of the system. Traditional methods require rigorous physical modeling and perform poorly for complex systems. In this paper, an end-to-end remaining useful life prediction method is proposed, which uses short-time Fourier transform (STFT) as preprocessing. Considering the time correlation of signal sequences, a long and short-term memory network is designed in CNN, incorporating the convolutional block attention module, and understanding the decision-making process of the network from the interpretability level. Experiments were carried out on the 2012PHM dataset and compared with other methods, and the results proved the effectiveness of the method. [ABSTRACT FROM AUTHOR]

The thermal imaging image of the Sustainable Development Science Satellite (SDGSAT-1) is mainly used for high-resolution observations of the ground width, due to the influence of blind elements and non-uniformity of the detector, and the system is a pendulum sweep imaging mode, resulting in fringed noise in the image. In this paper, a Fringing algorithm based on LRSID (low-rank-based single-image decomposition) algorithm is proposed, which can effectively remove the lateral and vertical fringe noise of the thermal imager and maintain the detail and clarity of the image. First, pretreatment of the obvious light and dark stripes then, based on LLSID algorithm, the vertical direction pinstripes and horizontal stripes are processed; finally, the fringed frequency band of the original image is replaced in the frequency domain with the image frequency domain processed by the LRSID algorithm, and then the Fourier inverse transformation is performed to obtain the final image. Using the method proposed in this paper, the simulated and actual SDGSAT-1 thermal imaging camera remote sensing stripes images are removed, and the visual and quantitative indicators are compared with the processing results of other algorithms, and the results show that the proposed algorithm has the best performance to remove the stripes, which can effectively remove horizontal and vertical fringes at the same time, and retain the detail and clarity of the image. [ABSTRACT FROM AUTHOR]