Magnitude Acceleration Research Articles

Performance-based liquefaction analysis and probabilistic liquefaction hazard mapping using CPT data within the Fraser River delta, Canada

Since the Fraser River delta is affected by multiple seismic sources, the selection of a single combination of magnitude and maximum ground acceleration to evaluate liquefaction initiation is challenging. In this study, we use a probabilistic seismic hazard analysis to account for all earthquake scenarios and also consider the liquefaction model uncertainty and soil resistance uncertainty using 787 CPT data across the study area. Hazard curves are generated for the factor of safety against liquefaction (FSL) and the amount of required soil improvement to prevent liquefaction (ΔqL) which provide a comprehensive assessment of liquefaction triggering. We derive the FSL and ΔqL values corresponding to return periods of 475 and 2475 years and generate the first probabilistic liquefaction hazard mapping for the region. Most of the lowland areas of Metro Vancouver correspond to low FSL and therefore soil improvement is needed for liquefaction mitigation.

Use of deep learning to improve seismic data quality analysis

Seismic data are influenced by various types of noise, which are typically categorized into two primary classes: anthropogenic and environmental. However, the detection of instrumental noise or malfunctioning stations also plays a crucial role in ensuring the data quality and the efficiency of a seismic network. The visual inspection of seismic spectral diagrams (e.g. power spectral density) enables us to identify issues that could potentially compromise data quality, thereby affecting subsequent calculations such as Magnitude or Peak Ground Acceleration (PGA). However, this process is time-consuming and demands significant human expertise due to the complexity of the diagrams, compounded by the sheer number of stations requiring examination. Therefore, in this paper, we explore the feasibility of transferring human expertise into an artificial intelligence system to create an automated system capable of rapidly performing such detection. More specifically, in the first part of this paper, we use Probability Density Function (PDF) diagrams, enabling an initial assessment of station performance via visual inspection. We describe this plot type and provide examples that reveal whether a station is functioning correctly or if technical issues exist. A table containing the main evaluation criteria is provided. In the second part of this paper, we demonstrate that these plots can serve as input for a neural network, allowing the development of the aforementioned automated system. Through extensive testing under various conditions, we have observed that the trained network consistently achieves an accuracy rate exceeding 85% across all four conductedtests. In the latest and most significant test, the achieved accuracy is approximately 87%.

Deep neural helmholtz operators for 3D elastic wave propagation and inversion

Summary Numerical simulations of seismic wave propagation in heterogeneous 3D media are central to investigating subsurface structures and understanding earthquake processes, yet are computationally expensive for large problems. This is particularly problematic for full waveform inversion, which typically involves numerous runs of the forward process. In machine learning there has been considerable recent work in the area of operator learning, with a new class of models called neural operators allowing for data-driven solutions to partial differential equations. Recent works in seismology have shown that when neural operators are adequately trained, they can significantly shorten the compute time for wave propagation. However, the memory required for the 3D time domain equations may be prohibitive. In this study, we show that these limitations can be overcome by solving the wave equations in the frequency domain, also known as the Helmholtz equations, since the solutions for a set of frequencies can be determined in parallel. The 3D Helmholtz neural operator is 40 times more memory-efficient than an equivalent time-domain version. We employ a Helmholtz neural operator for 2D and 3D elastic wave modeling, achieving two orders of magnitude acceleration compared to a baseline spectral element method. The neural operator accurately generalizes to variable velocity structures and can be evaluated on denser input meshes than used in the training simulations. We also show that when solving for wavefields strictly on the surface, the accuracy can be significantly improved via a graph neural operator layer. In leveraging automatic differentiation, the proposed method can serve as an alternative to the adjoint-state approach for 3D full-waveform inversion, reducing the computation time by a factor of 350.

Self-powered synchronous asymmetric voltage flip and charge extraction technique for piezoelectric energy harvesting

Abstract This paper presents a nonlinear interface circuit for piezoelectric vibration energy harvester (PEH) with Synchronous Asymmetric Voltage Flipping and Charge Extraction process, denoted as SAFCE. SAFCE flips the PEH voltage polarity at positive peak and completely extracting charge at negative peak through LC resonance. The harvested power is independent of load. In theory, the harvested power is 200 % of SECE and 780 % of best impedance-matched SEH due to the energy injection mechanism, which enhances the electromechanical coupling coefficient of PEH. Moreover, a self-powered SAFCE circuit without rectifier bridge is designed, which reduces power consumption and eliminates the need for external power sources. Experimental measurements are carried out to compare with SEH and SECE circuits under the condition of either constant displacement magnitude (0.5 mm) or constant external excitation acceleration (10 m/s2). The experimental results indicate that the power harvested by the SAFCE technique increased by 171 % compared with the SECE method and by 381 % compared with the best impedance-matched SEH method under the same conditions.

Threshold-based earthquake early warning for high-speed railways using deep learning

Earthquakes are disasters that threaten the operational safety of high-speed railways. To obtain reliable alerts for the earthquake monitoring and early warning systems of high-speed railways, based on magnitude and peak ground acceleration (PGA) thresholds (M = 5.5 and PGA = 40 cm/s2), an earthquake early warning (EEW) method for high-speed railways using deep learning is proposed. And the application of deep learning method in EEW for high-speed railway is explored. We design a single-station deep learning network architecture (named the CT architecture) by combining convolutional neural and transformer networks, and with that architecture, we train two separate models (CT-M and CT-PGA models) using the strong motion data recorded from the Kyoshin Network in Japan, which are used to predict whether the magnitude and PGA exceed the thresholds for issuing an alert. To verify the robustness of the method, we apply it to the M7.3 earthquake and M7.4 earthquake off the coast of Fukushima in 2021–2022. Results show that within 10 s after P-wave arrival, the accuracy of the alert reaches 90 %, and the average observed lead time reaches 18 s. The proposed method displays potential application on EEW systems for high-speed railways.

Adaptive cross tubal tensor approximation

In this paper, we propose a new adaptive cross algorithm for computing a low tubal rank approximation of third-order tensors, with less memory usage and lower computational complexity than the truncated tensor SVD (t-SVD). This makes it applicable for decomposing large-scale tensors. We conduct numerical experiments on synthetic and real-world datasets to confirm the efficiency and feasibility of the proposed algorithm. The simulation results show more than one order of magnitude acceleration in the computation of low tubal rank approximations for large-scale tensors. An application to pedestrian attribute recognition is also presented.

Deep mechanism reduction (DeePMR) method for fuel chemical kinetics

Fuel chemistry represents a typical complex system involving thousands of intermediate species and elementary reactions. Traditional mechanism reduction methods, such as sensitivity analysis and graph-based approaches, fail to explore global correlations of the sub-systems, thereby compromising their efficiency and accuracy. A novel machine learning-based approach called deep mechanism reduction (DeePMR) has been developed to address this issue. The current method transforms mechanism reduction into an optimization problem in the combinatorial space of chemical species while mitigating the curse of dimensionality inherent in the high-dimensional space. We propose an iterative sampling–training–predicting strategy combining deep neural networks with genetic algorithms to learn the landscape of the combinatorial space and locate the targeted subspace. Applying DeePMR to fuel chemistry mechanisms has led to much more compact mechanisms than traditional methods, including directed relation graph (DRG) or path flux analysis (PFA) methods, with three to four orders of magnitude acceleration in numerical simulation. In addition, reduced mechanisms by DeePMR indicate a principal-satellite formulation for constructing chemical reaction mechanisms, providing a straightforward yet effective alternative to hierarchy-based construction methods. The DeePMR method provides a general framework for model reduction across various fields.

Numerical Simulation as Laboratory Liquefaction Test Comparison for Geological Hazard Countermeasure: A Preliminary Study

Liquefaction is a secondary hazard due to earthquakes, which is one of the geological hazards. The impairment caused can damage infrastructure on a wide scale and even cause casualties. Several phenomenal liquefaction events have been recorded throughout the world, and one of the most phenomenal is the flow liquefaction incident in Palu, Sulawesi, Indonesia, in 2018. Research on liquefaction is still ongoing today. In this study, numerical simulations will be carried out as a comparison for laboratory liquefaction tests, which aim to understand the flow liquefaction mechanism better. Laboratory modeling is presented carefully in numerical modeling, namely by creating three (3) variations in sand density layers in a box with dimensions of 120 cm x 60 cm x 40 cm. The simulation uses the assumption that the soil is perfectly saturated by placing the groundwater level at an elevation of 0 cm. The acceleration applied to the box is varied in the range of 0.3 – 0.6g. Simulation shows that deformation increases with the bigger magnitude and Peak Ground Acceleration (PGA), while the duration is not too sensitive to the results. On the other hand, the area liquefied is affected by the duration, magnitude, and PGA value.

Unlocking the Key to Accelerating Convergence in the Discrete Velocity Method for Flows in the Near Continuous/Continuous Flow Regimes.

How to improve the computational efficiency of flow field simulations around irregular objects in near-continuum and continuum flow regimes has always been a challenge in the aerospace re-entry process. The discrete velocity method (DVM) is a commonly used algorithm for the discretized solutions of the Boltzmann-BGK model equation. However, the discretization of both physical and molecular velocity spaces in DVM can result in significant computational costs. This paper focuses on unlocking the key to accelerate the convergence in DVM calculations, thereby reducing the computational burden. Three versions of DVM are investigated: the semi-implicit DVM (DVM-I), fully implicit DVM (DVM-II), and fully implicit DVM with an inner iteration of the macroscopic governing equation (DVM-III). In order to achieve full implicit discretization of the collision term in the Boltzmann-BGK equation, it is necessary to solve the corresponding macroscopic governing equation in DVM-II and DVM-III. In DVM-III, an inner iterative process of the macroscopic governing equation is employed between two adjacent DVM steps, enabling a more accurate prediction of the equilibrium state for the full implicit discretization of the collision term. Fortunately, the computational cost of solving the macroscopic governing equation is significantly lower than that of the Boltzmann-BGK equation. This is primarily due to the smaller number of conservative variables in the macroscopic governing equation compared to the discrete velocity distribution functions in the Boltzmann-BGK equation. Our findings demonstrate that the fully implicit discretization of the collision term in the Boltzmann-BGK equation can accelerate DVM calculations by one order of magnitude in continuum and near-continuum flow regimes. Furthermore, the introduction of the inner iteration of the macroscopic governing equation provides an additional 1-2 orders of magnitude acceleration. Such advancements hold promise in providing a computational approach for simulating flows around irregular objects in near-space environments.

Accelerated fixed-point iterative reconstruction for fiber borescope imaging.

Computational imaging systems with embedded processing have potential advantages in power consumption, computing speed, and cost. However, common processors in embedded vision systems have limited computing capacity and low level of parallelism. The widely used iterative algorithms for image reconstruction rely on floating-point processors to ensure calculation precision, which require more computing resources than fixed-point processors. Here we present a regularized Landweber fixed-point iterative solver for image reconstruction, implemented on a field programmable gated array (FPGA). Compared with floating-point embedded uniprocessors, iterative solvers implemented on the fixed-point FPGA gain 1 to 2 orders of magnitude acceleration, while achieving the same reconstruction accuracy in comparable number of effective iterations. Specifically, we have demonstrated the proposed fixed-point iterative solver in fiber borescope image reconstruction, successfully correcting the artifacts introduced by the lenses and fiber bundle.

Physics-informed neural network with transfer learning (TL-PINN) based on domain similarity measure for prediction of nuclear reactor transients

Nuclear reactor safety and efficiency can be enhanced through the development of accurate and fast methods for prediction of reactor transient (RT) states. Physics informed neural networks (PINNs) leverage deep learning methods to provide an alternative approach to RT modeling. Applications of PINNs in monitoring of RTs for operator support requires near real-time model performance. However, as with all machine learning models, development of a PINN involves time-consuming model training. Here, we show that a transfer learning (TL-PINN) approach achieves significant performance gain, as measured by reduction of the number of iterations for model training. Using point kinetic equations (PKEs) model with six neutron precursor groups, constructed with experimental parameters of the Purdue University Reactor One (PUR-1) research reactor, we generated different RTs with experimentally relevant range of variables. The RTs were characterized using Hausdorff and Fréchet distance. We have demonstrated that pre-training TL-PINN on one RT results in up to two orders of magnitude acceleration in prediction of a different RT. The mean error for conventional PINN and TL-PINN models prediction of neutron densities is smaller than 1%. We have developed a correlation between TL-PINN performance acceleration and similarity measure of RTs, which can be used as a guide for application of TL-PINNs.

Analysis of liquefaction potential in the Malalayang beach area, North Sulawesi

North Sulawesi Province was locating in a tectonically active area, affected by subduction of the Sulawesi Sea plate, Sangihe, and Halmahera. These plates are still active to provide an earthquake threat. Several large earthquakes that came from various plates around Manado City have been recorded. A large earthquake that shakes sandy soil can produce liquefaction events. The Malalayang beach area, located in the West of Manado City a part of Malalayang Beach Area Arrangement Work Package and the Bunaken Village Ecotourism Arrangement, was chosen as the research object to analyse the possibility of liquefaction. The research begin with collect boring testing data, lab test data, and earthquake data such as the moment of magnitude and Peak Ground Acceleration (PGA). This study also uses various methods or attenuation equations to determine the peak ground acceleration from the history of earthquakes obtained from multiple sources. Based on the data obtained from the field and laboratory tests, data processing is carried out to determine the liquefaction potential. From the results of the Liquefaction Potential Index (LPI) calculation, it was found that the Malalayang Beach Area has the potential for the occurrence of the liquefaction phenomenon.

FDU-Net: Deep Learning-Based Three-Dimensional Diffuse Optical Image Reconstruction.

Near-infrared diffuse optical tomography (DOT) is a promising functional modality for breast cancer imaging; however, the clinical translation of DOT is hampered by technical limitations. Specifically, conventional finite element method (FEM)-based optical image reconstruction approaches are time-consuming and ineffective in recovering full lesion contrast. To address this, we developed a deep learning-based reconstruction model (FDU-Net) comprised of a Fully connected subnet, followed by a convolutional encoder-Decoder subnet, and a U-Net for fast, end-to-end 3D DOT image reconstruction. The FDU-Net was trained on digital phantoms that include randomly located singular spherical inclusions of various sizes and contrasts. Reconstruction performance was evaluated in 400 simulated cases with realistic noise profiles for the FDU-Net and conventional FEM approaches. Our results show that the overall quality of images reconstructed by FDU-Net is significantly improved compared to FEM-based methods and a previously proposed deep-learning network. Importantly, once trained, FDU-Net demonstrates substantially better capability to recover true inclusion contrast and location without using any inclusion information during reconstruction. The model was also generalizable to multi-focal and irregularly shaped inclusions unseen during training. Finally, FDU-Net, trained on simulated data, could successfully reconstruct a breast tumor from a real patient measurement. Overall, our deep learning-based approach demonstrates marked superiority over the conventional DOT image reconstruction methods while also offering over four orders of magnitude acceleration in computational time. Once adapted to the clinical breast imaging workflow, FDU-Net has the potential to provide real-time accurate lesion characterization by DOT to assist the clinical diagnosis and management of breast cancer.

High-speed photon correlation monitoring of amplified quantum noise by chaos using deep-learning balanced homodyne detection

Precision experimental determination of photon correlation requires massive amounts of data and extensive measurement time. We present a technique to monitor second-order photon correlation g(2)(0) of amplified quantum noise based on wideband balanced homodyne detection and deep-learning acceleration. The quantum noise is effectively amplified by an injection of weak chaotic laser, and the g(2)(0) of the amplified quantum noise is measured with a real-time sample rate of 1.4 GHz. We also exploit a photon correlation convolutional neural network accelerating correlation data using a few quadrature fluctuations to perform a parallel processing of g(2)(0) for various chaos injection intensities and effective bandwidths. The deep-learning method accelerates the g(2)(0) experimental acquisition with a high accuracy, estimating 6107 sets of photon correlation data with a mean square error of 0.002 in 22 s and achieving a three orders of magnitude acceleration in the data acquisition time. This technique contributes to a high-speed and precision coherence evaluation of entropy source in secure communication and quantum imaging.

Fast and accurate machine learning prediction of phonon scattering rates and lattice thermal conductivity

Lattice thermal conductivity is important for many applications, but experimental measurements or first principles calculations including three-phonon and four-phonon scattering are expensive or even unaffordable. Machine learning approaches that can achieve similar accuracy have been a long-standing open question. Despite recent progress, machine learning models using structural information as descriptors fall short of experimental or first principles accuracy. This study presents a machine learning approach that predicts phonon scattering rates and thermal conductivity with experimental and first principles accuracy. The success of our approach is enabled by mitigating computational challenges associated with the high skewness of phonon scattering rates and their complex contributions to the total thermal resistance. Transfer learning between different orders of phonon scattering can further improve the model performance. Our surrogates offer up to two orders of magnitude acceleration compared to first principles calculations and would enable large-scale thermal transport informatics.

Accelerated quantum Monte Carlo with probabilistic computers

Quantum Monte Carlo (QMC) techniques are widely used in a variety of scientific problems and much work has been dedicated to developing optimized algorithms that can accelerate QMC on standard processors (CPU). With the advent of various special purpose devices and domain specific hardware, it has become increasingly important to establish clear benchmarks of what improvements these technologies offer compared to existing technologies. In this paper, we demonstrate 2 to 3 orders of magnitude acceleration of a standard QMC algorithm using a specially designed digital processor, and a further 2 to 3 orders of magnitude by mapping it to a clockless analog processor. Our demonstration provides a roadmap for 5 to 6 orders of magnitude acceleration for a transverse field Ising model (TFIM) and could possibly be extended to other QMC models as well. The clockless analog hardware can be viewed as the classical counterpart of the quantum annealer and provides performance within a factor of < 10 of the latter. The convergence time for the clockless analog hardware scales with the number of qubits as ∼ N, improving the ∼ N2 scaling for CPU implementations, but appears worse than that reported for quantum annealers by D-Wave.

Artificial phosphatase upon premicellar nanoarchitectonics of lanthanum complexes with long-chained imidazole derivatives

Four novel long chain-containing tridentate imidazole derivatives (Ln, n = 1, 2, 3, 4) were synthesized for in situ formation of mononuclear lanthanum(III) complexes as artificial phosphodiesterases. These in-situ formed La(III) complexes (named LaLn) were used to catalyze the transesterification of 2-hydroxypropyl p-nitrophenyl phosphate (HPNP), a classic RNA model. Critical aggregation concentrations (CAC) were determined for the as-prepared tridentate imidazole derivatives as ligands and corresponding mixtures of equivalent ligand and La3+ ion with a mole rate of 1:1. It denotes that the introduction of La3+ ion increases the CAC values of imidazole derivatives by about 2 to 3 folds. Foaming test shows that the foam height is positively correlated with the length of hydrophobic chain. Transesterification of HPNP mediated by LaLn nanoarchitectonics indicates that the introducing of hydrophobic chain benefits rate enhancement, showing excess three orders of magnitude acceleration under physiological conditions (pH 7.0, 25 °C). Moreover, catalytic reactivities of these La(III) complexes increased along with the increase in chain length: LaL1 < LaL2 < LaL3 < LaL4, suggesting a positive correlation to hydrophobic chain length.

Joint ABS Deployment and TBS Antenna Downtilt Optimization for Coverage Maximization

In this letter, we consider an air-and-ground cooperative network, where several aerial base stations (ABS) help terrestrial base stations (TBS) for coverage enhancement. In this network, we first quantify the space-time coverage ratio (STCR) by fully considering the antenna models and the dynamic of the ABS, and then formulate a joint ABS deployment and TBS antenna downtilt optimization problem with the objective to maximize the STCR of the concerned area. The objective function involves many control variables and judgement operations, which make the problem very complex. To solve the problem effectively, we first adopt the genetic algorithm (GA). Using the solutions of the GA as training samples, we propose a deep neural network architecture to further reduce the computational time. Simulation results indicate that the proposed GA significantly improves the coverage ratio and the deep neural network (DNN) architecture achieves orders of magnitude acceleration in computational time with acceptable performance.

Examining the Ability of Two Actigraph Models to Detect and Discriminate Between Low Frequency Movements

Recently, there has been a notable shift to increasingly more sedentary lifestyles, yet it remains unclear whether inexpensive accelerometers, such as those contained in Actigraph devices, can provide consistent measurements of low magnitude accelerations. This study sought to explore the ability of two Actigraph models to differentiate low frequency oscillations, in reference to higher-end accelerometers with the idle sleep mode disabled (ISM) (Part 1) and enabled (Part 2) in a controlled environment. Eight GT9X, fifteen wGT3X-BT and two higher-end accelerometers (Triaxial ICP) were mounted to a 6-degree of freedom robot, which introduced frequencies ranging from 0.5-2.0 Hz (Part 1) and 0.5-4.0 Hz (Part 2). To compare the models, the minimum, maximum and range of outputs were calculated for each of the frequencies. Part 1 revealed that the Actigraph monitors were able to detect low frequency oscillations; the captured output was similar across the different Actigraph models but was significantly greater than the higher-end devices. Part 2 demonstrated that amplitudes greater than the described 40 m g (0.392 m/s <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> ) threshold were required for the Actigraph monitors to wake up. This study demonstrates that the GT9X and wGT3X-BT Actigraph accelerometers can detect low magnitude movements when the ISM is disabled.

Using Machine Learning to Greatly Accelerate Path Integral Ab Initio Molecular Dynamics.

Ab initio molecular dynamics (AIMD) has become one of the most popular and robust approaches for modeling complicated chemical, liquid, and material systems. However, the formidable computational cost often limits its widespread application in simulations of the largest-scale systems. The situation becomes even more severe in cases where the hydrogen nuclei may be better described as quantized particles using a path integral representation. Here, we present a computational approach that combines machine learning with recent advances in path integral contraction schemes, and we achieve a 2 orders of magnitude acceleration over direct path integral AIMD simulation while at the same time maintaining its accuracy.