Time Domain Research Articles

Optimizing tiny colorless feedback delay networks

A common bane of artificial reverberation algorithms is spectral coloration in the synthesized sound, typically manifesting as metallic ringing, leading to a degradation in the perceived sound quality. In delay network methods, coloration is more pronounced when fewer delay lines are used. This paper presents an optimization framework in which a tiny differentiable feedback delay network, with as few as four delay lines, is used to learn a set of parameters to iteratively reduce coloration. The parameters under optimization include the feedback matrix, as well as the input and output gains. The optimization objective is twofold: to maximize spectral flatness through a spectral loss while maintaining temporal density by penalizing sparseness in the parameter values. A favorable narrow distribution of modal excitation is achieved while maintaining the desired impulse response density. In a subjective assessment, the new method proves effective in reducing perceptual coloration of late reverberation. Compared to the author’s previous work, which serves as the baseline and utilizes a sparsity loss in the time domain, the proposed method achieves computational savings while maintaining performance. The effectiveness of this work is demonstrated through two application scenarios where smooth-sounding synthetic room impulse responses are obtained via the introduction of attenuation filters and an optimizable scattering feedback matrix.

Frequency and Time Domain Simulations of a 15 MW Floating Wind Turbine Integrating with Multiple Flap-Type WECs

This study integrates offshore wind power and wave power generation technologies to build a multi-energy complementary renewable energy system, which provides references for marine clean energy development and is highly consistent with the global sustainable development goals. The platform consists of a UMaine VolturnUS-S semi-submersible platform and a group of flap-type wave energy converters. A 15 MW wind turbine is installed on the platform. The hydrodynamic model is established using AQWA. Combined with the upper wind load, the fully coupled time domain model of the integrated power generation platform is constructed using the open-source software F2A. The main purpose is to optimize the parameters of the flap-type wave energy device through frequency domain hydrodynamic analysis and then explore the influence of the wave energy device on the platform under the combined action of regular waves and turbulent wind through a series of working conditions. The results show that when the PTO stiffness is 8 × 107 N·m/rad, the PTO damping takes the optimal damping and has a higher power generation capacity. Secondly, the coupled wave energy device induces minimal hydrodynamic interference between multiple bodies, resulting in negligible impact on the natural frequency of the wind-wave combined platform motion. Overall, the wave energy device can effectively suppress the freedom of shaking degree of the floating wind-wave combined platform.

Abstract A017: Enhanced efficacy of anti-mesothelin CAR T cells in ovarian cancer through drug combinations: Impact of treatment kinetics, co-stimulatory domain, and donor variability

Abstract Introduction: Clinical trials of mesothelin (MSLN) chimeric antigen receptor (CAR) T-cell therapy have shown limited efficacy in solid tumors like ovarian cancer (OVCA), emphasizing the need to enhance CAR T cell performance. This study evaluated whether combining MSLN-CAR T cells with drugs, either sequentially or concurrently, could achieve synergistic anti-tumor effects. Methods :MSLN-directed CAR T cells were generated using two constructs with CD28 (M28z) or 4-1BB (MBBz) co-stimulatory domains. A high-throughput drug screen of 528 compounds was conducted on MSLN-expressing OVCA cells in combination with CAR T cells, assessing three conditions: (1) simultaneous drug and CAR T exposure, (2) short drug pre-treatment of OVCA, and (3) extended drug pre-treatment before CAR T cell introduction. Drug-induced modulation of CAR T cell cytotoxicity was assessed by comparing CAR T cell-mediated killing, drug-only effects, and drug-CAR T cell combination effects on OVCA cells. Results: Synergy between drugs and CAR T cells depended on treatment timing and co-stimulatory domains. Chemotherapeutics enhanced CAR T cell cytotoxicity when administered as a pre-treatment, while epigenetic modulators and kinase inhibitors were more effective with concurrent CAR T cell application. Cisplatin and cytarabine significantly boosted MBBz CAR T activity, while idarubicin enhanced M28z CAR T cells. Certain agents, like vinflunine, gemcitabine, and the PARP inhibitor rucaparib, enhanced efficacy across both constructs. Donor-specific responses revealed two distinct donor clusters: one showed synergy with chemotherapeutics, and the other with targeted agents Conclusions: Combining drugs with CAR T cell therapy could improve the treatment efficacy in ovarian cancer. Pre-treatment with specific chemotherapeutics enhanced CAR T cell cytotoxicity, while simultaneous treatment with certain targeted therapies was also effective. The effects of CAR construct type and inter-donor variability highlight the potential for precision medicine approaches. These findings support further exploration of personalized therapeutic combinations to optimize CAR T cell efficacy in ovarian cancer and improve patient outcomes. Citation Format: Lucía Rico Pizarro, Thomas Poiret, Greta Gudoitytė, Olli Kallioniemi, Brinton Seashore-Ludlow, Isabelle Magalhaes, Tom Erkers. Enhanced efficacy of anti-mesothelin CAR T cells in ovarian cancer through drug combinations: Impact of treatment kinetics, co-stimulatory domain, and donor variability [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Functional and Genomic Precision Medicine in Cancer: Different Perspectives, Common Goals; 2025 Mar 11-13; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2025;85(5 Suppl):Abstract nr A017.

Autoregressive Moving Average Modelling of TESS Photometry of Massive Stars

Abstract Standard statistical time series analysis methods developed for observations regularly spaced in time are applied to measurements of 30 massive stars. Two TESS orbits of observations of each of the stars are analysed using this time domain methodology. It is found that the majority of light curves do not exhibit quasi-periodicities, i.e. the variability is fully aperiodic. The effective temperature and gravity can be used to predict with good accuracy the presence/absence of quasi-periodicities. Frequencies have a quadratic dependence on the effective temperature. The calculation of characteristic frequencies is also discussed. It is shown that the use of inappropriate spectral forms for this purpose may lead to biased results.

Unified force-based design approach for the seismic analysis and design of liquid storage tanks

Abstract Historical observations reveal that Liquid Storage Tanks (LST) have suffered significant earthquake-induced damages. The structural response of LST are sensitive to earthquakes due to dynamic fluid-tank interaction. Since designing with consideration of fluid-tank interaction in the time domain is complex, simplified calculation approaches have been developed to calculate base shear and overturning moments, but not pressure distributions. These approaches distinguish between flexible and rigid tanks, which is difficult to decide prior analysis. This paper presents a unified calculation concept that determines support reactions and pressure distributions independently of the tank’s stiffness by applying static equivalent loads. The research focuses on the distinction between rigid and flexible design approaches, review existing codes, their limitations, and challenges associated with relative acceleration response spectra. It scrutinizes varying definitions of impulsive components and superposition principles. A unified force-based design approach is suggested that integrates rigid and flexible design principles into a unified method. The approach uses standardized pressure curves of individual hydrodynamic pressure components, linked with absolute and relative spectral accelerations and appropriate superposition methods. The unified formulation is validated through a previously conducted experimental and numerical research on above-ground steel tank. The validation and application of the unified approach forms the basis for the new generation of Eurocode FprEN 1998-4 (2025), which is demonstrated for different tank geometries. The practical application is demonstrated on a squat and a slender tank using linear finite element model, and the results are compared with various approaches in the international standards and literature.

Methamphetamine alters the circadian oscillator and its couplings on multiple scales in Per1/2/3 knockout mice

Abstract Disruptions to circadian rhythms in mammals are associated with alterations in their physiological and mental states. Circadian rhythms are currently analysed in the time domain using approaches such as actograms, thus failing to appreciate their time-localised characteristics, time-varying nature and multiscale dynamics. In this study, we apply time-resolved analysis to investigate behavioural rhythms in Per1/2/3 knockout (KO) mice and their changes following methamphetamine administration, focusing on circadian (around 24 hours), low-frequency ultradian (around 7 hours), high-frequency ultradian (around 30 minutes) and circabidian (around 48 hours) oscillations. In the absence of methamphetamine, Per1/2/3 KO mice in constant darkness exhibited a dominant, approximately 7 hour oscillation. We demonstrate that methamphetamine exposure restores the circadian rhythm, although the frequency of the methamphetamine sensitive circadian oscillator (MASCO) varied considerably compared to the highly regular wild-type circadian rhythm. Additionally, methamphetamine increased multiscale activity and induced a circabidian oscillation in the Per1/2/3 KO mice. The information transfer between oscillatory modes, with frequencies around circadian, lowfrequency ultradian and high-frequency ultradian activity, due to their mutual couplings, was also investigated. For Per1/2/3 KO mice in constant darkness, the most prevalent coupling was between low and high-frequency ultradian activity. Following methamphetamine administration, the coupling between the circadian and high-frequency ultradian activity became dominant. In each case, the direction of information transfer was between the corresponding phases from the slower to faster oscillations. The time-varying nature of the circadian rhythm exhibited in the absence of Per1/2/3 genes and following methamphetamine administration may have profound implications for health and disease.

MRI-based radiomics for predicting pathological complete response after neoadjuvant chemoradiotherapy in locally advanced rectal cancer: a systematic review and meta-analysis

PurposeTo evaluate the value of MRI-based radiomics for predicting pathological complete response (pCR) after neoadjuvant chemoradiotherapy (NCRT) in patients with locally advanced rectal cancer (LARC) through a systematic review and meta-analysis.MethodsA systematic literature search was conducted in PubMed, Embase, Proquest, Cochrane Library, and Web of Science databases, covering studies up to July 1st, 2024, on the diagnostic accuracy of MRI radiomics for predicting pCR in LARC patients following NCRT. Two researchers independently evaluated and selected studies using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool and the Radiomics Quality Score (RQS) tool. A random-effects model was employed to calculate the pooled sensitivity, specificity, and diagnostic odds ratio (DOR) for MRI radiomics in predicting pCR. Meta-regression and subgroup analyses were performed to explore potential sources of heterogeneity. Statistical analyses were performed using RevMan 5.4, Stata 17.0, and Meta-Disc 1.4.ResultsA total of 35 studies involving 9,696 LARC patients were included in this meta-analysis. The average RQS score of the included studies was 13.91 (range 9.00-24.00), accounting for 38.64% of the total score. According to QUADAS-2, there were risks of bias in patient selection and flow and timing domain, though the overall quality of the studies was acceptable. MRI-based radiomics showed no significant threshold effect in predicting pCR (Spearman correlation coefficient=0.119, P=0.498) but exhibited high heterogeneity (I2≥50%). The pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio and DOR were 0.83, 0.82, 5.1, 0.23 and 27.22 respectively, with an area under the summary receiver operating characteristic (sROC) curve of 0.91. According to joint model analysis, publication year, country, multi-magnetic field strength, multi-MRI sequence, ROI structure, contour consistency, feature extraction software, and feature quantity after feature dimensionality reduction were potential sources of heterogeneity. Deeks’ funnel plot suggested no significant publication bias (P=0.69).ConclusionsMRI-based radiomics demonstrates high efficacy for predicting pCR in LARC patients following NCRT, holding significant promise for informing clinical decision-making processes and advancing individualized treatment in rectal cancer patients.Systematic review registrationhttps://www.crd.york.ac.uk/prospero/, identifier CRD42024611733.

Study of high-resolution analysis of Rayleigh waves on the basis of the Choi–Williams distribution

Rayleigh wave exploration is widely used in the field of engineering investigation, which can analyze the near surface velocity structure. However, the traditional Rayleigh wave exploration data processing method can only provide the average change of Rayleigh wave phase velocity with depth under the seismic line, but cannot obtain the horizontal change of Rayleigh wave phase velocity along the line direction, which brings great challenges to the detection of underground wave velocity anomalies. To solve these problems, the Choi‒Williams distribution is introduced to process Rayleigh waves, and the relationships among the Rayleigh wave time, frequency and amplitude are obtained. The frequency dispersion curve is subsequently calculated via similarity analysis. In this method, first, the Rayleigh wave is intercepted in the time domain. Second, the Choi‒Williams distribution of Rayleigh waves is calculated, and the time domain waveform of each frequency component of each Rayleigh wave is obtained. The time difference is subsequently calculated via similarity analysis, and the dispersion curve between two adjacent channels is obtained. Finally, the phase velocity image below the seismic arrangement is obtained via combination with the multichannel Rayleigh wave dispersion curve. When this method is applied to the detection of geological anomalies in engineering investigations, it can accurately reflect the spatial distribution position of the detection target and has a relatively high lateral resolution.

Triple Fano resonances induced by a square-elliptical cavity coupled with an MIM waveguide for high-sensitivity gas refractive index sensing

The proposed structure of a baffled metal–insulator–metal (MIM) waveguide coupled with a square cavity and a silver elliptical cylinder is designed to excite triple Fano resonances. The transmittance spectrum and electric field distribution at the resonance peaks produced by the structure are simulated using the finite difference time domain (FDTD) method. The simulation results of the transmittance spectrum are fitted by the multimode interference coupled-mode theory (MICMT). By analyzing the variations in the transmittance spectrum, the formation mechanism of the Fano resonances is investigated. The structure generates triple Fano resonances within the near-infrared spectrum, and its potential for gas refractive index sensing is investigated. The sensor demonstrated a sensitivity of up to 1942 nm/RIU and a figure of merit (FOM) of 1.10×105. These promising results suggest that the structure holds significant potential for applications in biochemical sensing.

Integration of BOTDA and φ-OTDR distributed fiber sensing for multiple parameters monitoring

Abstract In this article, the integration of the Brillouin optical time-domain analysis (BOTDA) and the phase-sensitive optical time domain reflectometry (φ-OTDR) distributed fiber sensing systems is discussed. By combining the two systems and sharing common instruments, the cost of the experiments is minimized, enabling the sensing of three parameters: temperature, strain, and vibration on the same 18.7 km fiber. In the BOTDA part, the focus is on temperature and strain analysis. Two 4 m dispersion shifted fibers (DSF) are used to simulate strain at different positions, causing the Brillouin frequency shift (BFS) to shift from 10.88 GHz to 10.52 GHz, allowing for strain sensing. The two 4 m single mode fibers and DSF are separately heated from room temperature to 200 °C, resulting in an R2 value of 0.97654 and 0.99958, respectively, indicating a good linear relationship. In the φ-OTDR section, vibration point frequencies were demodulated using an external interferometer system based on Mach–Zehnder interferometer (MZI). This article presents comprehensive experiments that demonstrate the ability to achieve multi-parameter sensing on a single fiber using a hybrid system, all within a low-cost framework.

Addressing the need for non-invasive lung assessment with time-resolved diffuse optics

Monitoring lungs functions is key for detecting several morbidities and pathologies. Photons in the 600–1,300 nm range might have the potential to reach lungs and provide compositional and functional information. Yet, few optical techniques have been challenged non-invasively so far. In this paper, we investigate the conditions to probe lungs using Time Domain Diffuse Optical Spectroscopy (TD-DOS). Counterintuitively, from Monte Carlo simulations we discovered that a higher absorption coefficient in the chest wall as compared to lungs increases sensitivity to deeper structures. In vivo measurements on the thorax of healthy volunteers during a forced breathing protocol, complemented with information on lung composition and previously evaluated in vivo spectra of porcine lung, suggest that this condition occurs above 1,100 nm. Multiple experimental setups were exploited to cover the 600–1,300 nm spectral range and test different source-detector distances (3–7 cm). All measurements exhibit oscillations consistent with the breathing rhythm, suggesting detection of lung expansion and compression. However, marked differences for different subjects and a complex dependence of the detected signal on the photon time-of-flight seem to allure to a non-trivial role of photon propagation through lungs, related–for instance–to the presence of alveoli and perhaps also to the overlying heterogeneous tissues. The unceasing development of time-resolved single-photon detectors with increasing performances above 1,000 nm, and a better understanding of lung optics–e.g., anomalous diffusion models–will help unravel the information from late, deep-travelling photons and lead to a novel photonic tool to probe the lungs non-invasively.

Traffic Flow Prediction Method Based on Deep Learning and Graph Neural Network

The continuous development of society is accelerating the process of urbanization and promoting the rapid construction of smart cities. As an important part of smart cities, many cities have begun to establish intelligent transportation systems (ITS). As the cornerstone of ITS, traffic flow prediction technology is crucial in trip prediction and urban traffic management. How to more effectively capture the spatiotemporal information present in traffic data is the fundamental problem of traffic flow prediction. However, existing methods have problems such as insufficient mining of spatiotemporal features and incomplete modeling of spatiotemporal relationships when performing traffic flow prediction tasks. Aiming at the problem that the existing study approaches lack dynamic modeling of traffic data, and RNN-based approaches have limitations in the capture of global information, this study proposes a self-attention-based spatial-temporal double graph convolutional networks for traffic flow forecasting (SASTDGCN). This method abstracts the influence between nodes in the time domain into a graph structure with relationships, and then uses relational graph convolutional neural networks (RGCN) for extraction of temporal features, so that road network nodes can directly obtain the traffic conditions of nodes at historical moments that are far apart, effectively addressing long-term forecasting. In addition, the model utilizes a module that is based on a dual-domain temporal and spatial self-concern mechanism for capturing the association between spatial and historical time steps of road nodes and generates a corresponding dynamic graph structure for the road network at each time step and the historical time steps of each node, successfully improving the model prediction accuracy by 13%.

A Fault Diagnosis Method for Oil Well Electrical Power Diagrams Based on Multidimensional Clustering Performance Evaluation

In oilfield extraction activities, traditional downhole condition monitoring is typically conducted using dynamometer cards to capture the dynamic changes in the load and displacement of the sucker rod. However, this method has severe limitations in terms of real-time performance and maintenance costs, making it difficult to meet the demands of modern extraction. To overcome these shortcomings, this paper proposes a novel fault detection method based on the analysis of motor power parameters. Through the dynamic mathematical modeling of the pumping unit system, we transform the indicator diagram of beam-pumping units into electric power diagrams and conduct an in-depth analysis of the characteristics of electric power diagrams under five typical operating conditions, revealing the impact of different working conditions on electric power. Compared to traditional methods, we introduce fourteen new features of the electrical parameters, encompassing multidimensional analyses in the time domain, frequency domain, and time-frequency domain, significantly enhancing the richness and accuracy of feature extraction. Additionally, we propose a new effectiveness evaluation method for the FCM clustering algorithm, integrating fuzzy membership degrees and the geometric structure of the dataset, overcoming the limitations of traditional clustering algorithms in terms of accuracy and the determination of the number of clusters. Through simulations and experiments on 10 UCI datasets, the proposed effectiveness function accurately evaluates the clustering results and determines the optimal number of clusters, significantly improving the performance of the clustering algorithm. Experimental results show that the fault diagnosis accuracy of our method reaches 98.4%, significantly outperforming traditional SVM and ELM methods. This high-precision diagnostic result validates the effectiveness of the method, enabling the efficient real-time monitoring of the working status of beam-pumping unit wells. In summary, the proposed method has significant advantages in real-time performance, diagnostic accuracy, and cost-effectiveness, solving the bottleneck problems of traditional methods and enhancing fault diagnosis capabilities in oilfield extraction processes.

Broad Learning System Based on Fractional Feature Optimization.

Broad learning system (BLS) have demonstrated excellent performance in terms of both speed and accuracy in tasks such as image classification. In BLS, the feature nodes predominantly utilize linear features, and sparse representation is mainly employed in the feature optimization component. The robustness of these features to different data needs to be improved. Although there are many improved algorithms for BLS in feature optimization, there is no improvement based on fractional calculus at present. This article proposes BLS-FC, a novel data classification and regression method that can seamlessly combine BLS and fractional calculation. Fractional calculus describes the properties of data between integer orders and has memory properties. Fractional Fourier transform (Frft) also has time domain and frequency domain information. First, Frft is added to the broad learning feature node extraction to enrich the node features, which is called BLS-Frft. Second, fractional calculus is integrated into the BLS-Frft sparse representation feature optimization, and the feature representation capability is enhanced by fractional differential memory. This part is called BLS-FS. Finally, in order to solve the problem of unstable features of random fractional order subspaces, a fractional order multiscale feature interaction based on BLS-Frft is proposed, which is called BLS-MF. Experimental results across various classification and regression datasets demonstrate the superior performance of the proposed method.

B-waves in noninvasive capacitance signal correlate with B-waves in ICP.

Analysis of B-waves in overnight intracranial pressure (ICP) recordings used to be an important element in the diagnosis of normal pressure hydrocephalus (NPH). Here, we tested the hypothesis that equivalents to B-waves can be detected and quantified in a noninvasively measured electric capacitance signal termed W. We measured ICP and W in a cohort of 15 patients with suspected diagnosis of NPH or spontaneous intracranial hypotension during infusion testing, identifying B-waves in both signals by wave-template matching in the time domain. We found very strong correlation between the duration of B-waves in ICP and W (R2 = 0.86, p < 10-6),and weak correlation between the average B-wave amplitudes in ICP and W (R = 0.34, p = 0.02). The concurrent presence of B-waves in the signals suggests that vasogenic activity of cerebral autoregulation is reflected in W.The weaker correlation of amplitudes may be attributed to W being an indirect measure of cranial volume composition, whereas ICP is a measure of pressure, with the two linked by the non-linear craniospinal pressure-volume relation that varies between patients. Analysis of the noninvasively acquired W signal should be evaluated as a triage tool for patients with NPH and other disorders characterized by reduced compliance.

Numerical and Experimental Investigation of rubbing Existence in the Context of a Rotating Machine Under Base excitation

Abstract This work presents an experimental investigation regarding the dynamic behavior of an onboard rotating machine in the case of disk-stator rubbing. The concept of onboard rotating machine has to do with the situation in which base excitations are considered in the mechanical modeling of the system. In the present study the system is composed of a disk mounted on a flexible shaft that is supported by rolling bearings at the ends of the shaft. The model is obtained from the classical Lagrange's equations and the finite element method, considering both the strain energy and kinetic energy of the shaft and the kinetic energies of the disk and the unbalance mass. The phenomenon of contact was included in the modeling of the onboard rotor. For this aim, two possible formulations for the contact effect were taken into consideration. The numerical results were compared with the experimental ones stemming from a test rig that was constructed to support the present research work. Different experimental tests were performed both in the frequency and the time domain, so that frequency response functions, orbits and unbalance responses were obtained. A good correspondence between numerical and experimental results was observed.

Depth domain time-lapse seismic simultaneous inversion based on accurate Zoeppritz equation

Abstract Time-lapse (TL) seismic technology has been widely adopted across various geophysical domains by quantifying reservoir dynamics through the analysis of discrepancies in time-lapse seismic data. Currently, established time-lapse seismic inversion methods are predominantly in the time domain. However, owing to advancements in depth migration technology, depth domain imaging is now recognized for its superior imaging accuracy. Relative to time domain imaging data, extracting requisite inversion outcomes directly from depth domain seismic data preserves more precise TL disparity information. In this article, we propose a method that directly performing pre-stack time-lapse seismic inversion in the depth domain to obtain time- lapse interpretation results in depth domain. In the article, we use a 1D point spread function to calculate seismic wavelets in the depth domain, and use the precise Zoeppritz equation to calculate the reflection coefficient. To enhance the stability of multi-parameter inversion, we establish an inversion objective function based on Bayesian framework and add low-wavenumber constraints and block constraints. Through synthetic test, we compared the separate inversion and differential inversion in the depth domain, results showed that time-lapse seismic simultaneous inversion is more suitable for depth domain applications. Finally, we validated the algorithm using real data and obtained high-quality application results.

Effect of Electrode Thickness on Capacitive and Self-Discharge Performance of Carbon-Based Supercapacitor

Supercapacitor is an energy storage device that has electrodes, electrolyte, separator and current collector components. In this study, activated carbon electrode material with turbostratic structure having the interlayer spacing (d002) of 3.550 Å, and with porosity having the specific surface area (SBET) of 685 m2 g-1 was used to prepare electrodes with the thickness of 0.1, 0.3 and 0.5 mm for symmetrical supercapacitor cells, respectively. Electrochemical tests show that the reduction in the electrodes thickness from the thickest electrodes to 0.1 mm causes the specifics capacitance, energy and power of the cells to increase by factors of 6, 5, and 3.5 times, respectively, to reach the values of 37 F kg-1, 5 W h kg-1, 114 W kg-1. This test also shows that the length of time domain in which the activation-controlled mechanisms that dominantly drives the voltage decay during self-discharge and the corresponding voltage decay rate, which are longer and higher for the cell (0.1 mm), respectively. Furthermore, the self-discharge data show the present of the time region of dominant diffusion-controlled mechanism for the cell (0.5 mm) but show the absence of such region for the other cells.

A Bearing Fault Diagnosis Method Based on Fusion of CNN-BiLSTM-Transformer and Cross-Attention

To address the limitations of traditional diagnostic models in achieving high accuracy in rolling bearing fault diagnosis and their inadequate extraction of vibration signal features across both frequency and time domains, this paper introduces a fault diagnosis approach that integrates multiple neural networks with a cross-attention mechanism. Initially, one-dimensional fault signals undergo processing through the Fast Fourier Transform (FFT) and Variational Mode Decomposition (VMD) techniques to obtain spectral information and time-domain features, respectively. The combined application of FFT and VMD facilitates the efficient extraction of multi-scale features from the signals. Subsequently, a CNN-Transformer network is utilized to capture the spatial characteristics of the pre-processed multi-scale fault signal features. This is followed by the deployment of a Bidirectional Long Short-Term Memory (BiLSTM) network to extract sequential information, with a particular emphasis on pivotal temporal features. Additionally, the integration of the Transformer network further enhances feature extraction and fusion capabilities. Ultimately, a cross-attention mechanism is implemented to seamlessly integrate time-domain and frequency-domain features, thereby bolstering the model's performance and generalization capacity for fault classification tasks. Experimental findings validate the proposed model's diagnostic effectiveness, achieving an accuracy and recall rate exceeding 99% in rolling bearing fault diagnosis.

A finite element analysis model for magnetomotive ultrasound elastometry magnet design with experimental validation

Objective. Magnetomotive ultrasound (MMUS) using magnetic nanoparticle contrast agents has shown promise for thrombosis imaging and quantitative elastometry via magnetomotive resonant acoustic spectroscopy (MRAS). Young's modulus measurements of smaller, stiffer thrombi require an MRAS system capable of generating forces at higher temporal frequencies. Solenoids with fewer turns, and thus less inductance, could improve high frequency performance, but the reduced force may compromise results. In this work, a computational model capable of assessing the effectiveness of MRAS elastometry magnet configurations is presented and validated.Approach. Finite element analysis (FEA) was used to model the force and inductance of MRAS systems. The simulations incorporated both solenoid electromagnets and permanent magnets in three-dimensional steady-state, frequency domain, and time domain studies.Main results. The model successfully predicted that a configuration in which permanent magnets were added to an existing MRAS system could be used to increase the force supplied. Accordingly, the displacement measured in a magnetically labeled validation phantom increased by a factor of 2.2 ± 0.3 when the force was predicted to increase by a factor of 2.2 ± 0.2. The model additionally identified a new solenoid configuration consisting of four smaller coils capable of providing sufficient force at higher driving frequencies.Significance. These results indicate two methods by which MRAS systems could be designed to deliver higher frequency magnetic forces without the need for experimental trial and error. Either the number of turns within each solenoid could be reduced while permanent magnets are added at precise locations, or a larger number of smaller solenoids could be used. These findings overcome a key challenge toward the goal of MMUS thrombosis elastometry, and simulation files are provided online for broader experimentation.