Search

With quaternion matrices and quaternion tensors being gradually used in the color image and color video processing, the block diagonalization of block circulant quaternion matrices has become a key issue in the establishment of T-product based methods for quaternion tensors. Out of this consideration, we aim at establishing a fast calculation approach for the block diagonalization of block circulant quaternion matrices with the help of the fast Fourier transform (FFT). We first show that the discrete Fourier matrix F p cannot diagonalize p × p circulant quaternion matrices, nor can the unitary quaternion matrices F p j and F p (1 + j) / 2 with j being an imaginary unit of quaternion algebra. Then we prove that the unitary octonion matrix F p p with p = l , il or (l + il) / 2 ( l , il being imaginary units of octonion algebra) can diagonalize a circulant quaternion matrix of size p × p , which further means that a block circulant quaternion matrix of size m p × n p can be block diagonalized at the cost of O (m n p log p) via the FFT. As one of applications, we give a fast algorithm to speed up the calculation of the T-product between m × n × p and n × s × p third-order quaternion tensors via FFTs, whose computational magnitude is almost 1/p of the original one. As another application, we propose an effective compression strategy for third-order quaternion tensors with a certain low-rankness. Simulations on the color image and color video compression demonstrate that our compression strategy with no QSVD involved, can achieve higher quality compression in terms of PSNR values at much less time costs, compared with the QSVD-based methods. [ABSTRACT FROM AUTHOR]

UNICEF reports that health workers frequently use the child's breath count to diagnose pneumonia in nations with weak health care systems (i.e., few doctors, limited access to laboratories, and chest x-rays). Counting a child's respiratory rate (RR) manually using an acute respiratory infection timer is a challenging task among health workers. The misclassification of an observed rate would lead to improper treatment. In the work, a low-cost noncontact automatic counter was proposed to estimate the RR using a Time of Flight (ToF) sensor. Here, the ToF sensor is pointed at the chest of the human body with clothes to capture chest wall expansion and contraction. This chest wall movement can be related to the inhalation and exhalation of breathing, which can be related to respiration activity. Clear breathing patterns were observed in the extracted respiration signal and standard signal processing techniques were employed to estimate the respiration rate. The device can measure the subject's respiration rate at a wide range of frequencies (slow, normal, and rapid), ranging from 4 to 100 breaths per minute (bpm). The single-pixel ToF sensor with a 940 nm wavelength is illumination-invariant in indoor applications and robust against clothes. The device's performance was assessed by comparing the attained real-time signals with the ground truth signals. The proposed device's estimated mean absolute error (MAE) has an error of less than 2 bpm for any frequency range. The device is computationally efficient, taking less than 36 ms to estimate and display the RR readings in the user interface. [ABSTRACT FROM AUTHOR]

In this paper, we consider solutions for discrete systems arising from multi-term time-fractional diffusion equations. Using discrete sine transform techniques, we find that all-at-once systems of such equations have a structure similar to that of diagonal-plus-Toeplitz matrices. We establish a generalized circulant approximate inverse preconditioner for the all-at-once systems. Through a detailed analysis of the preconditioned matrices, we show that the spectrum of the obtained preconditioned matrices is clustered around one. We give some numerical examples to demonstrate the effectiveness of the proposed preconditioner. [ABSTRACT FROM AUTHOR]

Human activity recognition (HAR) with wearable inertial sensors is a burgeoning field, propelled by advances in sensor technology. Deep learning methods for HAR have notably enhanced recognition accuracy in recent years. Nonetheless, the complexity of previous models often impedes their use in real-life scenarios, particularly in online applications. Addressing this gap, we introduce SenseMLP, a novel approach employing a multi-layer perceptron (MLP) neural network architecture. SenseMLP features three parallel MLP branches that independently process and integrate features across the time, channel, and frequency dimensions. This structure not only simplifies the model but also significantly reduces the number of required parameters compared to previous deep learning HAR frameworks. We conducted comprehensive evaluations of SenseMLP against benchmark HAR datasets, including PAMAP2, OPPORTUNITY, USC-HAD, and SKODA. Our findings demonstrate that SenseMLP not only achieves state-of-the-art performance in terms of accuracy but also boasts fewer parameters and lower floating-point operations per second. For further research and application in the field, the source code of SenseMLP is available at [ABSTRACT FROM AUTHOR]

In IoT applications, prediction models have fundamental challenges such as real-time processing, producing results with considerable/without delay, and taking action against pattern drift. While existing models can excel when data statistics remain relatively stable, real-time systems may encounter difficulties, particularly when confronted with dynamic shifts in data behavior. Analyzing data streams generated by different IoT applications and detecting complex pattern on the fly has become an open area of research. Complex event processing with adaptivity is a must to get desired features in such models. To address this issue, a comprehensive model for prediction has been proposed in this paper. It consists of two phases: (1) the basic model is constructed using historical data, (2) a fast Fourier transform-based adaptive support vector regression (FFT-ASVR) approach is proposed to predict events embedded in IoT data streams. FFT-ASVR predicts abnormal events by experiencing a change in data streams with real-time model updation. The performance of FFT-ASVR with a similar existing method SVM-RBF is presented using real-time traffic data of Madrid city. The proposed approach has significant improvement in terms of mean absolute percentage error (MAPE) for prediction, is adaptive in nature, and is also capable of handling the issue of pattern drift. [ABSTRACT FROM AUTHOR]

Surface plasmon resonance (SPR) sensor is a fast, label-free and real-time optical sensing technology. In this paper, an improved data-processing method of centroid algorithm was employed and the Fast Fourier Transform Algorithm (FFT) was also adopted to improve the standard deviation (SD) of resonance angle, a representative of sensing precision. The significant parameters in this study are the height ratio and selected spectrum width of centroid algorithm and the FFT cut-off frequency. To generate a RI change of sensing medium, pure water and 3 g/L NaCl solution were prepared for experiments. The SD was optimized in case of pure water, and then the RI sensitivity was also calculated for comparison. As results, two optimized regions were obtained, those are upper and under regions. The upper region is related with the selected spectrum width, and a low FFT cut-off frequency was suitable. The under region has a little change with the spectrum width and the FFT cut-off frequency. In experiments, the resolution of 2.11 × 10− 6 RIU was obtained with samples of ultrapure water and 3 g/L NaCl solution. Furthermore, the limit of detection for BSA molecule as low as 8.36 pM was achieved without any other noise reduction method. This study aims to provide a new design idea and theoretical support for the optimal design of SPR sensor, and has an important research significance and application prospect in the fields of environmental monitoring and biochemical sensing. [ABSTRACT FROM AUTHOR]

In this study, we propose a method for recommending appropriate combinations of stocks based on the waveforms of stock price changes. Many Japanese prefer to maintain stability when managing their assets. Specialized knowledge is required to invest in stocks while reducing risk. Hence, stock recommendation methods with different characteristics are required. Stock price movements can be captured as waveforms. Dynamic time warping (DTW), cross–correlation functions, and fast Fourier transforms (FFTs) are used to compare the features of the waveforms. A combination of stocks with different waveforms avoids the risk of simultaneous crashes. In the current experiment, one–year stock waveforms are used to obtain the recommended stock. The combined stocks selected using the DTW, cross–correlation functions, and FFT are shown to be suitable. [ABSTRACT FROM AUTHOR]

Based on contact data, the dynamics of the abundance (N) and biomass (B) of phytoplankton in 2013–2014 and the concentration of chlorophyll a (Cсhl) in 2000–2003 and 2008–2021 were analyzed using fast Fourier transform (FFT) under conditions of changing water temperature (T) in the vicinity of Sevastopol in the Black Sea. Estimates of the contribution made by the variability of the annual and semiannual harmonics of N, B, Cchl, T to the seasonal cycle amounted to more than 56% for all parameters. A significant relationship between B and Cchl (r < –0.83) at two stations indicates the aging of microalgae. The dominance of different groups of microalgae in the phytoplankton biomass has been detected. A typical period of 2–4 years is distinguished in the interannual variability of Cchl and T in different seasons. [ABSTRACT FROM AUTHOR]

Purpose: Probe-based confocal laser endomicroscopy (pCLE) enables visualization of cellular tissue morphology during surgical procedures. To capture high-quality pCLE images during tissue scanning, it is important to maintain close contact between the probe and the tissue, while also keeping the probe perpendicular to the tissue surface. Existing robotic pCLE tissue scanning systems, which rely on macroscopic vision, struggle to accurately place the probe at the optimal position on the tissue surface. As a result, the need arises for regression of longitudinal distance and orientation via endomicroscopic vision. Method: This paper introduces a novel method for automatically regressing the orientation between a pCLE probe and the tissue surface during robotic scanning, utilizing the fast Fourier vision transformer (FF-ViT) to extract local frequency representations and use them for probe orientation regression. Additionally, the FF-ViT incorporates a blur mapping attention (BMA) module to refine latent representations, which is combined with the pyramid angle regressor (PAR) to precisely estimate probe orientation. Result: A first of its kind dataset for pCLE probe-tissue orientation (pCLE-PTO) has been created. The performance evaluation demonstrates that our proposed network surpasses other top regression networks in accuracy, stability, and generalizability, while maintaining low computational complexity (1.8G FLOPs) and high inference speed (90 fps). Conclusion: The performance evaluation study verifies the clinical value of the proposed framework and its potential to be integrated into surgical robotic platforms for intraoperative tissue scanning. [ABSTRACT FROM AUTHOR]

Ptychography today is considered the most natural and effective method for approaching the diffraction limit of optical resolution. The principal scheme of a ptychoscope does not contain refractive or focusing elements and includes a coherent light source, a platform for moving (macroscopic) object, and a detector for registering radiation passed through or reflected by the object associated with a computer for processing diffraction patterns. In classical optics, the main task in achieving high spatial resolution is the correction and elimination of aberrations of optical systems, while in ptychography spatial resolution mainly depends on the reliability of registration and computer processing of diffraction patterns with large numerical apertures. The key idea of ptychography is to obtain an image by computer processing of overlapping diffraction patterns (scans). When moving the object it is assumed that the illuminating beam and the position of the detector remain unchanged. Ptychography is used in a wide range of radiation wavelengths from infrared to X-ray, and it is possible to do without imaging optics. In this work, the possibility of obtaining an amplitude-phase image of the surface relief of an object inclined to the incident one was studied. [ABSTRACT FROM AUTHOR]

We present a new term-structure model for commodity futures prices based on Trolle and Schwartz (2009), which we extend by incorporating seasonal stochastic volatility represented with two different sinusoidal expressions. We obtain a quasi-analytical representation of the characteristic function of the futures log-prices and closed-form expressions for standard European options' prices using the fast Fourier transform algorithm. We price plain vanilla options on the Henry Hub natural gas futures contracts, using our model and extant models. We obtain higher accuracy levels with our model than with the extant models. [ABSTRACT FROM AUTHOR]

As one of the main components of the Earth orientation parameters, short-term prediction of the geodetic polar motion series is crucial in the field of deep-space exploration, high-precision positioning, and timing services, which require high real-time performance. Additionally, its middle- and long-term prediction is equally important in climate forecasting and geodynamics research. In this study, we propose the combined BiLSTM+ARIMA model, which is based on bidirectional long- and short-term memory (BiLSTM) and autoregression integrated moving average (ARIMA). First, ensemble empirical mode decomposition (EEMD) is performed as a filter to decompose the polar motion time series to obtain low- and high-frequency signals. The EOP14 C04 time series provided by International Earth Rotation and Reference Systems Service and decomposed by EEMD includes low-frequency signals like the long-term trend, decadal oscillation, Chandler wobble, and prograde annual wobble, along with shorter-period high-frequency signals. Second, low- and high-frequency signals are predicted using BiLSTM and ARIMA models, respectively. Finally, the low- and high-frequency signal forecast components are reconstructed to obtain geodetic polar motion predictions. In middle- and long-term polar motion prediction, the results show that the proposed model can improve the prediction accuracy by up to 42% and 17%, respectively. This demonstrated that the BiLSTM+ARIMA model can effectively improve the accuracy of polar motion prediction. [ABSTRACT FROM AUTHOR]

Electrochemical impedance spectroscopy (EIS) is a measurement method widely used for non-destructive analysis and diagnostics in various electrochemical fields. From the measured dependence of the battery impedance on the frequency, it is possible to determine the parameters of various equivalent electrical circuit models of the battery. The conventional method of battery measurement using single-sine EIS is currently one of the most widely used methods for the analysis of lithium-ion batteries. However, its most significant disadvantage is the relatively long measurement time. For this reason, there is a growing demand for faster methods using fast-Fourier transform or pseudo-random sequences. A description of various EIS methods applications is provided in this paper. [ABSTRACT FROM AUTHOR]

Purpose: The current study focuses on measuring the level of vibration and noise produced by a direct injection diesel engine when running on a mixture of molybdenum trioxide (MoO3) nanoparticles and Calophyllum inophyllum biodiesel (B20). Methods: The nanoparticles were used in an amount of 75 ppm and mixed with a surfactant and dispersant. In addition, the prepared fuel samples were tested on a diesel engine to investigate the vibration and noise characteristics by changing the injection pressures to 200 bar, 220 bar and 240 bar. Results and conclusion: The addition of nanoparticles to B20 has significantly reduced vibration and noise, especially when using nanofuel enriched with dispersants. In addition, the increased pressure during fuel injection resulted in reduced vibration and noise levels. Using B20 + MoO3 75 ppm + Dispersant 75 ppm fuel blend gave the most favourable results regarding RMS velocity and noise levels compared to the base diesel fuel. The progressive reduction in vibration and noise levels was seen as a result of the increased injection pressure. At a pressure of 240 bar, the B20 + MoO3 75 ppm + dispersant 75 ppm has shown a lower vibration and noise intensity than diesel and other fuel samples. The RMS velocity of RMS noise of B20 + MoO3 75 ppm + dispersant 75 ppm are 0.591 m/s, 0.568 m/s, and 0.491 m/s at 200 bar, 220 bar and 240 bar while the RMS noise is 68.266 dB (A), 66.283 dB (A), and 65.857 dB (A) respectively. The most decrease in vibration and noise was 28.76% and 9.78%, respectively compared to regular diesel for B20 + MoO3 75 ppm + dispersant 75 ppm at 240 bar. Finally, the kurtosis and skewness of the vibration data are within the normal range, which indicates that the distribution corresponds to the ideal normal distribution. [ABSTRACT FROM AUTHOR]

Purpose: This work designs a procedure for structural damage assessment in beams under moving loads. The proposal is based on the respective alterations of hybrid vibrational factors with respect to the structural changes, whereby matching them for accurate predictions. These hybrid factors are extracted from the vibrational responses and then are employed for an artificial neural network (ANN) to assess the structural condition. Methods: An experiment is deployed to generate a databank for evaluation as well as ANN training. The maximal overlap discrete wavelet transform (MODWT) and the fast Fourier transform (FFT) are incorporated to split original signals into several power bands, which helps to exploit characteristics of individual vibrational modes: centroids and comparative powers of allocated spectra. The extracted hybrid factors are then fed into an ANN to predict the damage condition of the beam. The training process of this ANN uses a databank of extracted features from measured signals. Results: The paper successfully finds out the damage-sensitive feature through signal processing. Alternatively, the implemented experiment provides a rich databank for the ANN training, and the training performance reaches a high accuracy. Moreover, the testing results reveal the superior generalization of the ANN because of automatic predictions with high precision. This proves the contribution of the study that the exploited features are proper for structural assessment, and the trained ANN learns the correlation of vibration features and structural state well. Conclusions: The automation and effectiveness of the proposed approach are endorsed, which has high applicability and accuracy for damage assessment in beams. The proposed method is mainly devoted to identifying, quantifying, and localizing damage. [ABSTRACT FROM AUTHOR]

One of the most crucial components of music is accompaniment creation and chord arrangement is essential to accompanying development. A greater level of musical skill and a thorough knowledge of music theory are often required for competent accompaniment production. To organise and provide space for additional chords, a Bengio Nesterov Momentum optimizer to Bidirectional Encoder Representations from Transformers (BNM-BERT) intuitive computer approach is provided. To replace manual chord creation for accompaniment, the work has generated Optimised Intellectual Natural Language Processing Music Accompaniment. An Overlapping Segmentation based Independent Component Analysis (OS-ICA) technique, which ensuring the frame interval signals are continuous and smoothness is used to begin the procedure after segmentation. Then, theme signal feature extraction is performed using the Dual Fast Transform (DFT), which combines the Continues Wavelet Transform (CWT) and Fast Fourier Transform (FFT) for feature extraction from music audio signals. This results in coefficients for the timbral texture, tonal texture, Pitch class Profile (PCP), and rhythmic activity, among other features of musical information that are specific to the sound spectrum. Then, to enhance the musical accompaniment, very pertinent rhythmic signals are chosen employing HE instated Jellyfish inquiry advancement (HEinit-JSO). The unlabeled audio stream is labelled using average linkage based on hierarchical divisive clustering after the feature is chosen. Finally, automatic chord label creation is done using the BNM- BERT model. The examination reveals that the well-before algorithm outperforms the standards in terms of reconstructive probability and generating effectiveness. Additionally, it shows how cascading effects can be used for accompaniment improvement, texture generation from chord circumstances, and chord analysis. [ABSTRACT FROM AUTHOR]

Purpose: Data preprocessing is one of the key steps in any fault diagnosis process. The real data obtained from machines carry a lot of noise and inferred signals from other parts of the machine or environment. The intensity of this contamination varies with the sampling rate of data acquisition. To filter out these components and enhance the quality of the features generated from these data, several data preprocessing techniques are described in the literature. But the major concerns are the limitations of these techniques and the proper selection of sampling rates for data acquisition. Methods: This paper presents a comprehensive overview of parametric and non-parametric data preprocessing techniques for gearbox fault diagnosis and how these techniques preserve their properties under different sampling rates. Both analytically simulated signals and experimental signals are used in this work to check the effectiveness of these techniques at different sampling rates. Results and Conclusions: The obtained results clearly show that data preprocessed by a non-parametric filter contains significantly more information than data preprocessed by a parametric filter or without a filter. Even for a low (affordable) sampling rate, the non-parametric filter works well as compared to the parametric filter and with no filter. The proposed work has potential relevance in the industrial IoT for online condition monitoring of gearboxes. [ABSTRACT FROM AUTHOR]

In this paper, we consider a general notion of convolution. Let D be a finite domain and let D n be the set of n-length vectors (tuples) of D . Let f : D × D → D be a function and let ⊕ f be a coordinate-wise application of f. The f -Convolution of two functions g , h : D n → { - M , ... , M } is (g ⊛ f h) (v) : = ∑ v g , v h ∈ D n s.t. v = v g ⊕ f v h g (v g) · h (v h) for every v ∈ D n . This problem generalizes many fundamental convolutions such as Subset Convolution, XOR Product, Covering Product or Packing Product, etc. For arbitrary function f and domain D we can compute f -Convolution via brute-force enumeration in O ~ (| D | 2 n · polylog (M)) time. Our main result is an improvement over this naive algorithm. We show that f -Convolution can be computed exactly in O ~ ((c · | D | 2 ) n · polylog (M)) for constant c : = 3 / 4 when D has even cardinality. Our main observation is that a cyclic partition of a function f : D × D → D can be used to speed up the computation of f -Convolution, and we show that an appropriate cyclic partition exists for every f. Furthermore, we demonstrate that a single entry of the f -Convolution can be computed more efficiently. In this variant, we are given two functions g , h : D n → { - M , ... , M } alongside with a vector v ∈ D n and the task of the f -Query problem is to compute integer (g ⊛ f h) (v) . This is a generalization of the well-known Orthogonal Vectors problem. We show that f -Query can be computed in O ~ (| D | ω 2 n · polylog (M)) time, where ω ∈ [ 2 , 2.372) is the exponent of currently fastest matrix multiplication algorithm. [ABSTRACT FROM AUTHOR]

Background: Piston rings are one of the most vital and critical components of ICEs. The purpose of the piston rings is to provide effective sealing so that the heat from the piston can be transferred to the cylinder wall and oil from the crankcase should be kept away from the combustion chamber. The wear of piston rings directly affects the efficiency and performance of the engine. As a result, to diagnose piston rings faults, an effective condition monitoring and fault diagnosis methodology must be adopted. Purpose: In this study, an attempt has been carried out to diagnose piston ring scuffing faults using vibration signature analysis. Methods: An accelerometer sensor was used to collect time domain vibration signals from the healthy and all faulty engine operating conditions. Spectral feature analysis techniques such as the fast Fourier transform (FFT), power spectrum density (PSD), and empirical mode decomposition (EMD) were used to post-process the acquired signals. Further, a neural network technique known as multi-layer perceptron (MLP) was employed to develop the model for the classification of piston ring scuffing faults. In the end, the outcomes of signal processing and machine learning techniques were evaluated for both engine operating conditions. Results: The results showed clear shreds of fault diagnostic information using FFT, PSD, and EMD analyses whereas the MLP technique provided the best neural network architecture with the least mean squared error value. Conclusion: The results obtained using the signal processing techniques can be considered as a reference for the diagnosis of various engine component faults viz. gudgeon pin, gears, bearings, camshaft, etc. Further, the classification models created with these signals may be utilized in the automobile sector to forecast faults in the previously stated engine components. [ABSTRACT FROM AUTHOR]

Dataflow architecture is an alternative to traditional von Neumann computing architecture. However, known variants of dataflow architecture have a range of serious problems with no effective solutions up to the present day. This paper represents Hybrid Recurrent Signal Processor's (HRSP) hardware verification results. It describes HRSP's register transfer level model implementing its architectural specification and hardware prototype on the HAN Pilot Platform development board with Intel Arria10 field-programmable gate array. HRSP consists of a von Neumann master processor on a control layer and a recurrent dataflow unit on an operational layer. Dataflow unit includes four computing cores. HRSP's hardware model combines either software or hardware implementation of the control processor and the hardware model of the operational layer. Testing the HRSP's hardware prototype on the development board using an isolated word recognizer (IWR) as a typical data processing application has proven that the hardware model is bit-exact with both HRSP's imitation model and the original IWR C++ model. The HRSP's hardware prototype's achieved performance ensures IWR's operation in real-time mode on the development board. It is slightly better than the performance of the TMSC55x (Texas Instruments) digital signal processor. Verification of the HRSP's hardware implementation on synthetic tests showed that its average performance is 5% higher than the performance of the DSP TMSC55x digital signal processor. The results of the proposed optimization of hardware support for Fast Fourier Transform (FFT) in HRSP prove that such an optimization speeds up the FFT calculation, significantly reduces the capsule size, reduces the required hardware resources and simplifies FFT scaling. [ABSTRACT FROM AUTHOR]

This paper reports on a comparative experimental study of the current waveforms accompanying the propagation of creeping discharges over different liquid/pressboard interfaces under both negative and positive lightning impulse voltages; the investigated insulating liquids are mineral oil (MO), a synthetic ester (SE), a natural ester (NE) and an extra virgin olive oil (OO). Two different arrangements are used; in the first one, the point electrode was positioned in a perpendicular direction to the pressboard, and in the second one, the pressboard was positioned parallel to the axis of the point electrode. Due to the similarity of the recorded creeping discharge current shapes, it is hard to differentiate between the currents measured for each type of interface with time domain approaches. To circumvent this difficulty and identify the specific feature(s) associated to each type of interface, two techniques are used to analyze the signal of the discharge currents, namely discrete wavelet and fast Fourier-transform methods. It is shown that: (1) for the first arrangement, there is a particular spectral signature for each type of liquid/pressboard interface, and (2) for the second arrangement, the obtained results are in accordance to those obtained for the first arrangement only in some cases. [ABSTRACT FROM AUTHOR]

Although deep learning makes great success in increasing the accuracy for long sequence time-series forecasting, its complex neural network structure, which comprises many different types of layers and each layer containing hundreds and thousands of neurons, challenges the computing and memory capability of embedded platforms. This paper proposes a lightweight and efficient neural network called TTFNet, which forecasts long time series using three types of features (i.e., the Trend, Temporal attention, and Frequency attention) extracted from raw time series. In TTFNet, we perform a pooling operation on the historical data in a recent time window to extract a general trend, use a multi-layer perceptron to discover the temporal correlation between data as temporal attention, and apply the fast Fourier transforms on data to obtain frequency information as frequency attention. Each feature is separately extracted from its neural network branch with an output result, and we weight the three results to generate the final prediction while optimal weights are learnt. Also, the three prediction results can run in parallel since they are independent from one another. The experimental results show that the proposed method reduces the memory overhead and runtime by 62% and 81% of the five counterpart methods on average while achieving comparable performance. [ABSTRACT FROM AUTHOR]

This research paper introduces a novel approach using dominant frequency analysis to diagnose inner race faults in rolling ball bearings of three-phase induction motor. The main objective of the proposed scheme is to identify damaged bearings by analyzing their characteristic frequency components within a specific time interval in segmented signal. This work has been carried out on vibration signal data provided by Bearing Center Case Reserve Western University (CWRU), USA. In this study, IR007, IR014 and IR021 bearing defects are analyzed by Frequency Statistical Analysis in MATLAB. Mean and standard deviation of dominant frequencies are computed from the recorded vibration signals after dividing the signal into multiple segments of equal length. It is observed that both frequency mean, and standard deviation have been found to be highly sensitive with variations of motor speed and connected load. Therefore, motor speed is also studied to calculate the statistical parameters. The test results suggest that the proposed scheme provides comprehensive information about fault analysis through vibration data and could potentially aid researchers in fault analysis using the CRWU datasets. Overall, this paper presents a promising approach to diagnose the faults in the inner race of rolling ball bearings using frequency domain analysis and statistical parameters. [ABSTRACT FROM AUTHOR]

Due to the complexity of underwater scenes, underwater imaging is affected by the absorption and scattering of light through the water medium, resulting in degradation such as color casts, low contrast, and blurring. Many existing deep learning-based methods often fail to consider the impact of image degradation on different color channels, resulting in limited performance. In this work, we propose a progressive attention network based on RGB and HSV color spaces to improve the visual quality of underwater images. Specifically, we first transform the raw image into HSV color space and form RGB and HSV branches, respectively. We separate the color channels in each branch and employ the designed FFT-based aggregated residual dense module to focus on spatial and frequency domain features. Second, we feed the learned features of each channel on the two branches into the triple-color channel-wise attention module separately to balance the color distribution of different channels. Finally, we combine the results from two branches to obtain a high-quality image. Extensive qualitative and quantitative experimental results show that our model outperforms the compared methods. To evaluate the effectiveness of the components of our model, we conduct a series of ablation experiments. [ABSTRACT FROM AUTHOR]

Based on a brief review of forward algorithms for the computation of topographic gravitational and magnetic effects, including spatial, spectral and hybrid-domain algorithms working in either Cartesian or spherical coordinate systems, we introduce a new algorithm, namely the CP-FFT algorithm, for fast computation of terrain-induced gravitational and magnetic effects on arbitrary undulating surfaces. The CP-FFT algorithm, working in the hybrid spatial-spectral domain, is based on a combination of CANDECOMP/PARAFAC (CP) tensor decomposition of gravitational integral kernels and 2D Fast Fourier Transform (FFT) evaluation of discrete convolutions. By replacing the binomial expansion in classical FFT-based terrain correction algorithms using CP decomposition, convergence of the outer-zone computation can be achieved with significantly reduced inner-zone radius. Additionally, a Gaussian quadrature mass line model is introduced to accelerate the computation of the inner zone effect. We validate our algorithm by computing the gravitational potential, the gravitational vector, the gravity gradient tensor, and magnetic fields caused by densely-sampled topographic and bathymetric digital elevation models of selected mountainous areas around the globe. Both constant and variable density/magnetization models, with computation surfaces on, above and below the topography are considered. Comparisons between our new method and space-domain rigorous solutions show that with modeling errors well below existing instrumentation error levels, the calculation speed is accelerated thousands of times in all numerical tests. We release a set of open-source code written in MATLAB language to meet the needs of geodesists and geophysicists in related fields to carry out more efficiently topographic modeling in Cartesian coordinates under planar approximation. [ABSTRACT FROM AUTHOR]

In this paper we compare three different orthogonal systems in L 2 (R) which can be used in the construction of a spectral method for solving the semi-classically scaled time dependent Schrödinger equation on the real line, specifically, stretched Fourier functions, Hermite functions and Malmquist–Takenaka functions. All three have banded skew-Hermitian differentiation matrices, which greatly simplifies their implementation in a spectral method, while ensuring that the numerical solution is unitary—this is essential in order to respect the Born interpretation in quantum mechanics and, as a byproduct, ensures numerical stability with respect to the L 2 (R) norm. We derive asymptotic approximations of the coefficients for a wave packet in each of these bases, which are extremely accurate in the high frequency regime. We show that the Malmquist–Takenaka basis is superior, in a practical sense, to the more commonly used Hermite functions and stretched Fourier expansions for approximating wave packets. [ABSTRACT FROM AUTHOR]

The electrochemical behavior of disordered systems, such as high-entropy alloys, is a stochastic random process. To accurately predict and analyze the behavior of such systems under operating conditions, it is necessary to use new computational and experimental methods along with classical electrochemical methods. Using equimolar rare-earth alloys GdTbDyHoSc and GdTbDyHoY as an example, we demonstrate the efficiency of using fast Fourier transform and wavelet analysis to estimate the electrochemical behavior of stochastic systems. The time series of potential fluctuations of alloy samples are measured in 0.01 M NaCl solution within 12 h at a current density of 0.2–0.5 mA/cm2. Fast Fourier transform analysis of the obtained time series shows that the slope of the logarithm of spectral power density to the logarithm of frequency increases with the current density. In particular, coefficient β changes from –1.93 to –1.77 for a GdTbDyHoY sample and from –1.46 to –1.35 for a GdTbDyHoSc sample. In addition, wavelet analysis is used to process the time series obtained for both alloys at current densities from 0.2 to 0.5 mA/cm2. To illustrate the intensity of the electrochemical dissolution of the alloy surface, we construct scalograms for the obtained time series. The scalograms are used to calculate the global energy spectra distributed over frequency ranges and the total energies of the systems under study. The GdTbDyHoY alloy exhibits higher total energies as compared to the GdTbDyHoSc alloy. The total energy for the GdTbDyHoY alloy increases from 0.97 to 2.03 kV2 when the current density increases from 0.2 to 0.5 mA/cm2. For the GdTbDyHoSc alloy, the total energy increases from 0.50 to 0.84 kV2. Fast Fourier transform and wavelet analysis are found to be effective tools for understanding the electrochemical behavior of locally disordered chemical systems, such as high-entropy GdTbDyHoSc and GdTbDyHoY alloys, in addition to classical electrochemical methods. [ABSTRACT FROM AUTHOR]

Energy sources and mechanisms supplying energy to the oscillating system of an ejector pulsejet engine that supports gas acoustic self-oscillations, are researched. The study is carried out using experimental, numerical, and analytical methods. [ABSTRACT FROM AUTHOR]

Different omics profiles, depending on the underlying technology, encompass measurements of several hundred to several thousand molecules in a biological sample or a cell. This study develops upon the concept of "omics imagification" as a process of transforming a vector representing these numerical measurements into an image with a one-to-one relationship with the corresponding sample. The proposed imagification process transforms a high-dimensional vector of molecular measurements into a two-dimensional RGB image to enable holistic molecular representation of a biological sample and to improve the classification of different biological phenotypes using automated image recognition methods in computer vision. A transformed image represents 2D coordinates of molecules in a neighbour-embedded space representing molecular abundance and gene intensity. The proposed method was applied to a single-cell RNA sequencing (scRNA-seq) data to "imagify" gene expression profiles of individual cells. Our results show that a simple convolutional neural network trained on single-cell transcriptomics images accurately classifies diverse cell types outperforming the best-performing scRNA-seq classifiers such as support vector machine and random forest. [ABSTRACT FROM AUTHOR]

This paper examines the compensation of capacitive currents using arc suppression reactors with magnetization. A method is developed for automatic tuning of an arc suppression reactor for improving the accuracy and reliability of the results of measurements of the capacitance of the zero sequence loop. These problems are solved by measuring and recording the frequency spectra of the neutral voltage using the fast Fourier transform before and after perturbation of the zero sequence loop by a reference current of short duration through a signal winding of the arc suppression reactor. After this, the difference in the obtained frequency spectra is isolated and the proper frequency of the zero sequence circuit is determined from the maximum amplitude corresponding to the frequency of the free oscillations. The found frequency for a known inductance of the arc suppression reactor is used to determine the capacitive impedance of the network and the required magnetization current. [ABSTRACT FROM AUTHOR]

The time data series of weather stations are a source of information for floods. The study of the previous wintertime series allows knowing the behavior of the variables and the result that will be applied to analysis and simulation models that feed variables such as flow and level of a study area. One of the most common problems is the acquisition and transmission of data from weather stations due to atypical values and lost data; this generates difficulties in the simulation process. Consequently, it is necessary to propose a numerical strategy to solve this problem. The data source for this study is a real database where these problems are presented with different variables of weather. This study is based on comparing three methods of time series analysis to evaluate a multivariable process offline. For the development of the study, we applied a method based on the discrete Fourier transform (DFT), and we contrasted it with methods such as the average and linear regression without uncertainty parameters to complete missing data. The proposed methodology entails statistical values, outlier detection, and the application of the DFT. The application of DFT allows the time series completion, based on its ability to manage various gap sizes and replace missing values. In sum, DFT led to low error percentages for all the time series (1% average). This percentage reflects what would have likely been the shape or pattern of the time series behavior in the absence of misleading outliers and missing data. [ABSTRACT FROM AUTHOR]

In this paper, a numerical method to solve the multi-dimensional spatial fractional Allen–Cahn equations has been investigated. After semi-discretizating the equations, a system of nonlinear ordinary differential equations with a Toeplitz structure is induced. We propose to split the Toeplitz matrix into the sum of a circulant matrix and a skew-circulant matrix, and apply the Strang splitting method. Such a two-level Strang splitting method will reduce the computational complexity to O (q log q) . Moreover, it preserves not only the discrete maximum principle unconditionally but also second-order convergence as well. By introducing a new modified energy formula, the energy dissipation property can be guaranteed. Finally, some numerical experiments are conducted to confirm the theories we put forward. [ABSTRACT FROM AUTHOR]

We address two computational issues common to open-population N-mixture models, hidden integer-valued autoregressive models, and some hidden Markov models. The first issue is computation time, which can be dramatically improved through the use of a fast Fourier transform. The second issue is tractability of the model likelihood function for large numbers of hidden states, which can be solved by improving numerical stability of calculations. As an illustrative example, we detail the application of these methods to the open-population N-mixture models. We compare computational efficiency and precision between these methods and standard methods employed by state-of-the-art ecological software. We show faster computing times (a ∼ 6 to ∼ 30 times speed improvement for population size upper bounds of 500 and 1000, respectively) over state-of-the-art ecological software for N-mixture models. We also apply our methods to compute the size of a large elk population using an N-mixture model and show that while our methods converge, previous software cannot produce estimates due to numerical issues. These solutions can be applied to many ecological models to improve precision when logs of sums exist in the likelihood function and to improve computational efficiency when convolutions are present in the likelihood function. Supplementary materials accompanying this paper appear online. [ABSTRACT FROM AUTHOR]

The harmonic inversion (HI) problem in nuclear magnetic resonance spectroscopy (NMR) is conventionally considered by means of parameter estimations. It consists of extracting the fundamental pairs of complex frequencies and amplitudes from the encoded time signals. This problem is linear in the amplitudes and nonlinear in the frequencies that are entrenched in the complex damped exponentials (harmonics) within the time signal. Nonlinear problems are usually solved approximately by some suitable linearization procedures. However, with the equidistantly sampled time signals, the HI problem can be linearized exactly. The solution is obtained by relying exclusively upon linear algebra, the workhorse of computer science. The fast Padé transform (FPT) can solve the HI problem. The exact analytical solution is obtained uniquely for time signals with at most four complex harmonics (four metabolites in a sample). Moreover, using only the computer linear algebra, the unique numerical solutions, within machine accuracy (the machine epsilon), is obtained for any level of complexity of the chemical composition in the specimen from which the time signals are encoded. The complex frequencies in the fundamental harmonics are recovered by rooting the secular or characteristic polynomial through the equivalent linear operation, which solves the extremely sparse Hessenberg or companion matrix eigenvalue problem. The complex amplitudes are obtained analytically as a closed formula by employing the Cauchy residue calculus. From the frequencies and amplitudes, the components are built and their sum gives the total shape spectrum or envelope. The component spectra in the magnitude mode are described quantitatively by the found peak positions, widths and heights of all the physical resonances. The key question is whether the same components and their said quantifiers can be reconstructed by shape estimations alone. This is uniquely possible with the derivative fast Padé transform (dFPT) applied as a nonparametric processor (shape estimator) at the onset of the analysis. In the end, this signal analyzer can determine all the true components from the input nonparametric envelope. In other words, it can quantify the input time signal. Its performance is presently illustrated utilizing the time signals encoded at a high-field proton NMR spectrometer. The scanned samples are for ovarian cyst fluid from two patients, one histopathologically diagnosed as having a benign lesion and the other with a malignant lesion. These findings are presently correlated with the NMR reconstruction results from the Padé-based solution of the HI problem. Special attention is paid to the citrate metabolites in the benign and malignant samples. The goal of this focus is to see whether the citrates could also be considered as cancer biomarkers as they are now for prostate (low in cancerous, high in normal or benign tissue). Cancer biomarkers are metabolites whose concentration levels can help discriminate between benign and malignant lesions. [ABSTRACT FROM AUTHOR]

Graph Neural Network (GNN) architecture is a state-of-the-art model, which can obtain complete node embedding features and rich data information by aggregating the information of nodes and neighbors. Therefore, GNNs are widely used in electronic shopping, drug discovery (especially for the treatment of COVID-19) and other fields, promoting the explosive development of machine learning. However, user interaction, data sharing and circulation are highly sensitive to privacy, and centralized storage can lead to data isolation. Therefore, Federated Learning with high efficiency and strong security and privacy enhancement technology based on secure aggregation can improve the security dilemma faced by GNN. In this paper, we propose an Efficient Secure Aggregation for Federated Graph Neural Network(ESA-FedGNN), which can efficiently reduce the cost of communication and avoid computational redundancy while ensuring data privacy. Firstly, a novel secret sharing scheme based on numerical analysis is proposed, which employs Fast Fourier Transform to improve the computational power of the neural network in sharing phase, and leverages Newton Interpolation method to deal with the disconnection and loss of the client in reconstruction phase. Secondly, a regular graph embedding based on geometric distribution is proposed, which optimizes the aggregation speed by using data parallelism. Finally, a double mask is adopted to ensure privacy and prevent malicious adversaries from stealing model parameters. We achieve O (log N log (log N)) improvements compared to O N 2 in state-of-the-art works. This research helps to provide security solutions related to the practical development and application of privacy-preserving graph neural network technology. [ABSTRACT FROM AUTHOR]

Identifying potential threats concealed within the baggage is of prime concern for the security staff. Many researchers have developed frameworks that can automatically detect baggage threats from security X-ray scans. However, to the best of our knowledge, all of these frameworks require extensive training efforts on large-scale and well-annotated datasets, which are hard to procure in the real world, especially for the rarely seen contraband items. This paper presents a novel unsupervised anomaly instance segmentation framework that recognizes baggage threats, in X-ray scans, as anomalies without requiring any ground truth labels. Furthermore, thanks to its stylization capacity, the framework is trained only once, and at the inference stage, it detects and extracts contraband items regardless of their scanner specifications. Our one-staged approach initially learns to reconstruct normal baggage content via an encoder–decoder network utilizing a proposed stylization loss function. The model subsequently identifies the abnormal regions by analyzing the disparities within the original and the reconstructed scans. The anomalous regions are then clustered and post-processed to fit a bounding box for their localization. In addition, an optional classifier can also be appended with the proposed framework to recognize the categories of these extracted anomalies. A thorough evaluation of the proposed system on four public baggage X-ray datasets, without any re-training, demonstrates that it achieves competitive performance as compared to the conventional fully supervised methods (i.e., the mean average precision score of 0.7941 on SIXray, 0.8591 on GDXray, 0.7483 on OPIXray, and 0.5439 on COMPASS-XP dataset) while outperforming state-of-the-art semi-supervised and unsupervised baggage threat detection frameworks by 67.37%, 32.32%, 47.19%, and 45.81% in terms of F1 score across SIXray, GDXray, OPIXray, and COMPASS-XP datasets, respectively. [ABSTRACT FROM AUTHOR]

Accurate fault diagnosis and prognosis can significantly reduce maintenance costs, increase the safety and availability of engineering systems that have become increasingly complex. It has been observed that very limited researches have been reported on fault diagnosis where multi-component degradation are presented. This is essentially a challenging Complex Systems problem where features multiple components interacting simultaneously and nonlinearly with each other and its environment on multiple levels. Even the degradation of a single component can lead to a misidentification of the fault severity level. This paper introduces a new test rig to simulate the multi-component degradation of the aircraft fuel system. A machine learning-based data analytical approach based on the classification of clustering features from both time and frequency domains is proposed. The scope of this framework is the identification of the location and severity of not only the system fault but also the multi-component degradation. The results illustrate that (a) the fault can be detected with accuracy > 99%; (b) the severity of fault can be identified with an accuracy of almost 100%; (c) the degradation level can be successfully identified with the R-square value > 0.9. [ABSTRACT FROM AUTHOR]

Computational homogenization is the gold standard for concurrent multi-scale simulations (e.g., FE2) in scale-bridging applications. Often the simulations are based on experimental and synthetic material microstructures represented by high-resolution 3D image data. The computational complexity of simulations operating on such voxel data is distinct. The inability of voxelized 3D geometries to capture smooth material interfaces accurately, along with the necessity for complexity reduction, has motivated a special local coarse-graining technique called composite voxels (Kabel et al. Comput Methods Appl Mech Eng 294: 168–188, 2015). They condense multiple fine-scale voxels into a single voxel, whose constitutive model is derived from the laminate theory. Our contribution generalizes composite voxels towards composite boxels (ComBo) that are non-equiaxed, a feature that can pay off for materials with a preferred direction such as pseudo-uni-directional fiber composites. A novel image-based normal detection algorithm is devised which (i) allows for boxels in the firsts place and (ii) reduces the error in the phase-averaged stresses by around 30% against the orientation cf. Kabel et al. (Comput Methods Appl Mech Eng 294: 168–188, 2015) even for equiaxed voxels. Further, the use of ComBo for finite strain simulations is studied in detail. An efficient and robust implementation is proposed, featuring an essential selective back-projection algorithm preventing physically inadmissible states. Various examples show the efficiency of ComBo against the original proposal by Kabel et al. (Comput Methods Appl Mech Eng 294: 168–188, 2015) and the proposed algorithmic enhancements for nonlinear mechanical problems. The general usability is emphasized by examining various Fast Fourier Transform (FFT) based solvers, including a detailed description of the Doubly-Fine Material Grid (DFMG) for finite strains. All of the studied schemes benefit from the ComBo discretization. [ABSTRACT FROM AUTHOR]

In this paper, we present a very fast Monte Carlo scheme for additive processes: the computational time is of the same order of magnitude of standard algorithms for simulating Brownian motions. We analyze in detail numerical error sources and propose a technique that reduces the two major sources of error. We also compare our results with a benchmark method: the jump simulation with Gaussian approximation. We show an application to additive normal tempered stable processes, a class of additive processes that calibrates “exactly” the implied volatility surface. Numerical results are relevant. This fast algorithm is also an accurate tool for pricing path-dependent discretely-monitoring options with errors of one basis point or below. [ABSTRACT FROM AUTHOR]

The invertibility of a Toeplitz matrix can be assessed based on the solvability of two standard equations. The inverse of the nonsingular Toeplitz matrix can then be represented as the sum of products of circulant and skew-circulant (CS) matrices. In this note, we provide a new structured perturbation analysis for the CS representation of Toeplitz inversion and the new upper bound is just half as large as the existing upper bound proposed by Wu et al. (Numer Linear Algebra Appl 22(4):777–792, 2015) and Feng et al. (East Asian J Appl Math 5(2):160–175, 2015). Meanwhile, some practical issues and numerical experiments involving the numerical solutions of fractional partial differential equations are reported to support our theoretical findings. [ABSTRACT FROM AUTHOR]

Fast and accurate calculation of gravitational effects on a regional or global scale with complex density environment is a critical issue in gravitational forward modelling. Most existing significant developments with tessroid-based modelling are limited to homogeneous density models or polynomial ones of a limited order. Moreover, the total gravitational effects of tesseroids are often calculated by pure summation in these methods, which makes the calculation extremely time-consuming. A new efficient and accurate method based on tesseroids with a polynomial density up to an arbitrary order in depth is developed for 3D large-scale gravitational forward modelling. The method divides the source region into a number of tesseroids, and the density in each tesseroid is assumed to be a polynomial function of arbitrary degree. To guarantee the computational accuracy and efficiency, two key points are involved: (1) the volume Newton’s integral is decomposed into a one-dimensional integral with a polynomial density in the radial direction, for which a simple analytical recursive formula is derived for efficient calculation, and a surface integral over the horizontal directions evaluated by the Gauss–Legendre quadrature (GLQ) combined with a 2D adaptive discretization strategy; (2) a fast and flexible discrete convolution algorithm based on 1D fast Fourier transform (FFT) and a general Toepritz form of weight coefficient matrices is adopted in the longitudinal dimension to speed up the computation of the cumulative contributions from all tesseroids. Numerical examples show that the gravitational fields predicted by the new method have a good agreement with the corresponding analytical solutions for spherical shell models with both polynomial and non-polynomial density variations in depth. Compared with the 3D GLQ methods, the new algorithm is computationally more accurate and efficient. The calculation time is significantly reduced by 3 orders of magnitude as compared with the traditional 3D GLQ methods. Application of the new algorithm in the global crustal CRUST1.0 model further verifies its reliability and practicability in real cases. The proposed method will provide a powerful numerical tool for large-scale gravity modelling and also an efficient forward engine for inversion and continuation problems. [ABSTRACT FROM AUTHOR]

A high-accuracy economical iterative method is proposed for calculating the potential and the strength of the electric field in a three-dimensional inhomogeneous spatially periodic dielectric placed in an initially uniform electric field. The idea underlying the algorithm is that the potential is represented as a sum of a linear function and a spatially periodic correction, which can be expressed as an expansion in eigenfunctions of the Laplace operator that satisfy the appropriate periodicity conditions. The fast Fourier transform is used for an efficient numerical implementation of the proposed algorithm. [ABSTRACT FROM AUTHOR]

The concrete biological shield of light water reactors is exposed to neutron and gamma irradiation throughout its lifetime, which results in the long-term degradation of the concrete's mechanical properties. Under neutron irradiation, the concrete's aggregates are subjected to radiation-induced volumetric expansion (RIVE), which strongly depends on the mineral content of the aggregate and exhibits the largest expansion in silicate-bearing minerals. In this work, the authors used the fast Fourier transform-based code Microstructure-Oriented Scientific Analysis of Irradiated Concrete (MOSAIC) in 2D to model the expansion of five different aggregates provided by the Japan Concrete Aging Management Program (JCAMP). Comparable rock specimens were irradiated at the JEEP-II test reactor. The model uses realistic aggregate microstructure reconstruction based on high-resolution characterization images. The model accounts for anisotropic RIVE, thermal expansion, and the associated initiation and propagation of damage. The RIVE models are calibrated based on expansion data in the literature. The authors assume that damage occurs exclusively at interfaces between the particles that compose an aggregate and that these interfaces also exhibit swelling. Using a micromechanical model, the evolution of Young's modulus with RIVE is calculated for each aggregate and compared with Russian irradiation data. The modeled linear expansion agrees well with the experimentally measured expansion. The model also predicts that anisotropic RIVE and thermal expansion result in an earlier onset of damage with neutron fluence than in the isotropic case. [ABSTRACT FROM AUTHOR]

Analyzing massive spatial datasets using a Gaussian process model poses computational challenges. This is a problem prevailing heavily in applications such as environmental modeling, ecology, forestry and environmental health. We present a novel approximate inference methodology that uses profile likelihood and Krylov subspace methods to estimate the spatial covariance parameters and makes spatial predictions with uncertainty quantification for point-referenced spatial data. “Kryging” combines Kriging and Krylov subspace methods and applies for both observations on regular grid and irregularly spaced observations, and for any Gaussian process with a stationary isotropic (and certain geometrically anisotropic) covariance function, including the popular Matérn covariance family. We make use of the block Toeplitz structure with Toeplitz blocks of the covariance matrix and use fast Fourier transform methods to bypass the computational and memory bottlenecks of approximating log-determinant and matrix-vector products. We perform extensive simulation studies to show the effectiveness of our model by varying sample sizes, spatial parameter values and sampling designs. A real data application is also performed on a dataset consisting of land surface temperature readings taken by the MODIS satellite. Compared to existing methods, the proposed method performs satisfactorily with much less computation time and better scalability. [ABSTRACT FROM AUTHOR]

This contribution details the effective and accurate determination of the time-domain harmonic state estimation (TDHSE) of power networks interconnecting wind generation sources. The TDHSE method obtains the state variables of power networks and the system inputs. The algorithm uses a limited number of measuring devices, which can be contaminated with noise and/or gross errors. The wind generation source model represents the dynamic operation of a wind energy conversion system. It is based on a type-4 wind turbine and a direct-drive permanent magnet synchronous generator with a full scale back-to-back power converter. The wind generation source assumes wind fluctuations and changes in the angular speed control drive to estimate its dynamic behaviour and the effect on the power network. Case studies are considered for the analysis of a three-phase power network, i.e. with harmonic injections, with a wind generation source and with a wind farm. The TDHSE results of the analysed case studies are validated through direct comparison against the PSCAD/EMTDC® solution, obtaining a close agreement between the TDHSE and simulator responses. [ABSTRACT FROM AUTHOR]

Apples are susceptible to various types of damage during the production process. Such damage not only affects the appearance and edibility of the apples but also can result in the infection of healthy apples, leading to secondary economic losses. Therefore, it is crucial to properly handle damaged apples and re-sort them to enhance their utilization value and optimize resource use. To quickly and accurately identify apple damage and perform sorting in real time, addressing the resource limitations of mobile devices and the difficulty of extracting deep network image features, this study proposes a lightweight real-time apple damage classification network, Fast Fourier Transform Channel Attention (FFTCA)-YOLOv8n-cls. The FFTCA module focuses on the frequency domain feature information of images in deep networks, enhancing the network’s feature extraction capabilities. Additionally, it integrates Convolutional Block Attention Module (CBAM) and Distribution Shifting Convolution to capture channel and spatial information of images in shallow networks and accelerate network inference. Finally, FFTCA-YOLOv8n-cls is compared with typical lightweight classification networks. Experimental results show that this network has better classification accuracy and faster inference speed. Specifically, the FFTCA-YOLOv8n-cls network is only 0.601 MB in size, achieving a classification accuracy of 96.03%, a recall of 96.08%, and an F1-score of 96.05%, demonstrating its feasibility in real-time apple damage sorting. Moreover, this study applies the network to sorting robots, completing backend inference on servers and real-time inference on embedded devices to adapt to different working environments, achieving real-time sorting of damaged apples and validating the network’s application effectiveness. [ABSTRACT FROM AUTHOR]

The topic of this study is in vitro proton magnetic resonance spectroscopy (MRS). The theme is on theoretical analysis of time signals encoded at a high magnetic field 14.1T, using a Bruker spectrometer, operating at a Larmor frequency of 600 MHz. The samples, dissolved in a D 2 O buffer, are from histopathologically analyzed ovarian cyst fluid from two patients. The benign and malignant diagnoses were serous cystadenoma and serous cystadenocarcinoma, respectively. It is of vital clinical importance to determine whether certain specific patterns, inferred from the analyzed/interpreted MRS data could be correlated with this and similar histopathologic findings for other patients. Encoded time signals contain the fingerprint of the examined sample, its metabolic content. Therefore, to detect the sought patterns from MRS data, the salient characteristics of a malignant tumor, implied by the diagnostically most relevant metabolites (including recognized cancer biomarkers, e.g. lactic acids, cholines,...), need to be unambiguously identified by their significant departures from the associated control data of benign biomaterial, ovarian cyst fluid (serous cystadenoma) in the diagnostic problem under the present consideration. Such identifications are unfeasible by visualization in the domain of encoding (time domain). A direct inspection of the graphed waveforms of an encoded time signal would give no clue about its structure nor about the sample content. However, merely visualizing the plots of the equivalent, information-preserving spectral lineshape profiles in the frequency domain would make transparent at least some of clinically useful, discernible features of MRS data, a number of resonances assignable to the known and unknown metabolites. For instance, the size of each resonance (peak area) is proportional to the concentration of the given metabolite. This is a key quantitative measure, which could help differentiate a malignant from a benign specimen by reference to the normal standards. A number of metabolites (choline, alanine, lactate, threonine, β -hydroxybuturate, valine, isolecine, leucine,...) have substantially different concentrations in the malignant compared with normal samples. Time signals can be processed by two substantially different categories of mathematical transforms, shape and parameter estimators. The former processors are alternatively called nonparametric estimators. They have been employed for envelopes in our recent study on this problem, which will presently be addressed with the prime focus on reconstructions of the corresponding components. Components and envelopes are partial and total shape spectra, respectively. The sum of all the component lineshapes (one per metabolite) yields the envelope nondegenerate spectrum representation of the entire sample. Presently, a deeper diagnostically valid insight is gained about the metabolic content of the scanned sample through the reported exact component spectra. The employed parameter estimators are the high-resolution, noise-suppressing nonderivative and derivative fast Padé transforms. Detailed are several critical achievements by the parametric Padé processing of direct clinical relevance. Importantly, all the accomplishments are shared by the nonparametric derivative Padé estimations. Three examples are highlighted here as follows. Confirmation of our recent nonparametric derivative detection of an unassigned metabolite (a singlet peak) co-resonating with free choline near chemical shift 3.19 ppm (parts per million). Therein, with the nonderivative envelope, only one compound peak usually appears and is conventionally assigned to a free choline singlet. However, such an oversight would yield about twice larger value of the true concentration of this key cancer biomarker. The concentration level of another cancer biomarker (lactate) is also overestimated by any nonparametric nonderivative envelope. In sharp contrast, the parametric nonderivative Padé estimation unequivocally detects six usually invisible resonances (assignable to other metabolites) beneath the lactate doublet, around chemical shift 1.41 ppm. At least two of the strongest among these invisible six resonances can be also identified in the nonparametric fourth derivative Padé envelope. Regularization of the spectral compound for the water residual (4.71 ppm), which deforms the neighboring resonance lineshapes and impacts adversely on the concentration assessments of other nearby metabolites. This is accomplished by the fourth derivative envelope (coincident with the components) whose narrowing of the widths, cutting off the long tails and the background flattening generate a quantifiable singlet of water. This can serve as a reliable calibration reference resonance. After such a localization, no distortion appears around water, so that even very near 4.71 ppm, several smaller resonances are detected (assignable to a multiplet of nitrogen acetyl asparate), totally invisible in the nonparametric nonderivative envelope. [ABSTRACT FROM AUTHOR]

Populations are ageing and the healthcare costs are increasing accordingly. Humanoid Robots (HRs) perform basic but crucial health checkups such as the heart rate can efficiently meet the healthcare demands. This paper develops a 9-stage heart rate estimation algorithm and implements it to an HR. The 9-stages cover the recognition of the face with the Viola–Jones algorithm, determination of the facial regions with the geometric-based facial distance measurement technique, extraction of the forehead and cheek regions, tracking of these facial regions with the Hierarchical Multi Resolution algorithm, decomposition of the facial regions in the Red–Green–Blue (RGB) color channels, averaging and normalization of the RGB color data, elimination of the artifacts with the Adaptive Independent Component Analysis (ICA) technique, calculating the power spectrum of the data with the Fast Fourier Transform (FFT) technique, and finally determining the peaks inside the threshold reflecting the human heart rate boundaries. One of the key contributions of this paper is building and incorporating the Hierarchical Multi Resolution technique in the heart rate estimation algorithm to eliminate the deteriorating effects of the human and camera motions. A further contribution of this paper is generating a rule-based approach to discard the effects of the sudden movements. These two contributions have noticeably improved the accuracy of the heart rate estimation algorithm in the dynamic environments. The algorithm has been assessed extensively with 5 different experimental scenarios consisting of 33 conditions. [ABSTRACT FROM AUTHOR]

Background: Bearing faults in centrifugal pumps usually result in periodically fluctuating torque, which leads to flow instabilities, pressure pulsation, excessive heating, and even pump failure. Purpose: This paper presents a theoretical approach for modeling and analyzing the impact of some typical bearing faults on the dynamic characteristics of a monoblock centrifugal pump. Method: The governing equations of the flow are solved by the method of characteristics, while the dynamic parameters of the asynchronous motor are predicted by adopting the d-q axes theory. Bearing faults are modeled as periodic square-wave pulses injected into the shaft of the pump. Result: The studies of the simulation results show that bearing fault has no significant impact on the pump H-Q curve but it has a large negative effect on the efficiency curve. Additionally, bearing failure in centrifugal pumps can induce head oscillation and flow rate fluctuation. Conclusion: In this work, the influence of bearing anomalies on the performance of centrifugal pumps was analyzed using a numerical method. The model analysis indicated that bearing faults in centrifugal pumps can lead to remarkable performance degradation, severe vibrations, and significant flow instabilities. [ABSTRACT FROM AUTHOR]

Photonic Bandgap crystals are avidly studied because of their application of optical waveguides, optical diodes, defect mode photonic lasers, to name a few. This paper presents a comparative study of 14 bandgap crystals [13 defective and 1 without defect]. The design, simulation, and classification-based analysis of these defective crystals obtained by modifying dielectric arrangement in photonic bandgap lattice (PBG) lattice is carried out. In addition, the study of the dielectric structure of rods in the air is carried out concerning photonic bandgap for a 9 × 10μm wafer size. The discrete pulse components in real data, imaginary data, and intensity data were obtained using Fast Fourier Transforms and were converted to a clean dataset. With context to defective Photonic Crystal analysis and design explored the predictive and generative models for data-driven approaches. Within the predictive modeling framework, Microsoft AzureML is used to give classification performance using five different algorithms. The relative effectiveness of these algorithms has been studied. The created data have been classified using Multiclass Artificial Neural Network, Multiclass Decision Jungle, Multiclass Logistic Regression, Multiclass Random Forest, and 2-Clsaas Support Vector Machine classifiers. The Multiclass Decision Jungle and Multiclass Random Forest exhibit the maximum accuracy of 91.01% and 91.20%, respectively, while the other three algorithms give the classification accuracy of 87.5%. [ABSTRACT FROM AUTHOR]