Search

In this paper, we report a new algorithm for creating an adaptive basis set in the Multiconfigurational Ehrenfest (MCE) method, which is termed Full Cloning (FC), and test it together with the existing Multiple Cloning (MC) using the spin-boson model at zero-temperature as a benchmark. The zero-temperature spin-boson regime is a common hurdle in the development of methods that seek to model quantum dynamics. Two versions of MCE exist. We demonstrate that MC is vital for the convergence of MCE version 2 (MCEv2). The first version (MCEv1) converges much better than MCEv2, but FC improves its convergence in a few cases where it is hard to converge it with the help of a reasonably small size of the basis set. [ABSTRACT FROM AUTHOR]

Understanding the dynamics of bubble growth and collapse during gas–liquid two-phase transient flow is crucial for minimizing pipeline wear and maintaining the normal operation of the pipeline system. This paper focuses on obtaining the bubble trajectory, movement velocity, and volume change of gas bubbles. The bubble images during the transient flow are first denoised and enhanced, then the moving target detection algorithm and watershed segmentation algorithm are used to segment the bubbles, and the contour reconstruction of the bubbles is performed using fitting to extract the relevant parameters. The movement trajectories of the bubbles are then traced by analyzing the trajectories and velocities of growth and collapse of the first bubbles during the transient flow, specifically when the flow rate of the hydraulic pipeline is 0.75 m/s. Finally, the changes in gas content during the transient flow in the pipeline under different initial flow rates are compared between the algorithm proposed here and the simulation results. [ABSTRACT FROM AUTHOR]

Security issues and privacy issues are serious problems facing today's society, especially in image security, where privacy protection plays a pivotal role. To improve the security of images, we propose an image encryption algorithm based on optical chaos and Rubik's cube matrix in this paper. First, optical chaos is generated by constructing an optical device model. Second, in the image encryption algorithm, optical chaos and Rubik's cube matrix are used to encrypt the image at the bit level for the first time, and a "U" type encryption method is designed, and different "U" type encryption schemes are selected to encrypt the image for the second time. Finally, the "four-way diffusion" algorithm is used to diffuse the encrypted image, which further improves the security of the image. The computer simulations and security analysis results both confirm that ciphertext images can resist various common attack means, such as statistical attacks, differential attacks, and brute force attacks. In this paper, the proposed algorithm of decimal conversion, "U" encryption, and "quadrangle diffusion" makes the pixel value and pixel position change greatly, and the ciphertext image loses the original features of the plaintext image, which shows that the algorithm has good security performance and is suitable for image encryptions. [ABSTRACT FROM AUTHOR]

Within the context of fewest-switch surface hopping (FSSH) dynamics, one often wishes to remove the angular component of the derivative coupling between states J and K . In a previous set of papers, Shu et al. [J. Phys. Chem. Lett. 11, 1135–1140 (2020)] posited one approach for such a removal based on direct projection, while we isolated a second approach by constructing and differentiating a rotationally invariant basis. Unfortunately, neither approach was able to demonstrate a one-electron operator O ̂ whose matrix element J O ̂ K was the angular component of the derivative coupling. Here, we show that a one-electron operator can, in fact, be constructed efficiently in a semi-local fashion. The present results yield physical insight into designing new surface hopping algorithms and are of immediate use for FSSH calculations. [ABSTRACT FROM AUTHOR]

In this paper we present the Markovian Multiagent Monte-Carlo Second Order Self-Consistent Field Algorithm (M3-SOSCF). This algorithm provides a highly reliable methodology for converging SCF calculations in single-reference methods using a modified differential evolution approach. Additionally, M3 is embarrassingly parallel and modular in regards to Newton–Raphson subroutines. We show that M3 is able to surpass contemporary SOSCFs in reliability, which is illustrated by a benchmark employing poor initial guesses and a second benchmark with SCF calculations which face difficulties using standard SCF algorithms. Furthermore, we analyse inherent properties of M3 and show that in addition to its robustness and efficiency, it is more user-friendly than current SOSCFs. [ABSTRACT FROM AUTHOR]

The template matching method has been widely utilized in the defect detection of wafer surfaces. However, the traditional matching approaches are limited by illumination, noise, and deformation, which cannot meet the requirements of accuracy and robustness. In this paper, a novel multiple targets localization method, named Normalized Cross-correlation Adaptive Variable Step-Size Dynamic Template (NCC-AVSSDT) matching, is proposed to improve the accuracy and efficiency of image localization, which combines the advantages of NCC and AVSSDT. The AVSSDT method is utilized to dynamically adjust the scanning step size based on the NCC matching coefficients. This approach optimizes the scanning process, accelerating convergence toward the optimal matching position. Experimental results verify the accuracy and robustness of the proposed method under different conditions, especially when dealing with rotational variations and variations in noise textures. Therefore, NCC-AVSSDT can be used to perform multiple targets localization of chip image in nearly real-time. Three experiment types were used for comprehensive evaluations, including multiple targets, noise, and rotation angles. Experimental results show that NCC-AVSSDT is much better than the sequential similarity detection algorithm and mean absolute deviation methods in terms of multiple targets (0.667 vs 0.811 s, 0.832 s) and success rate (100% vs 35%, 20%). [ABSTRACT FROM AUTHOR]

Digital fringe projection profilometry based on phase-shifting technology is a reliable method for complex shape measurement, and the phase is one of the most important factors affecting measurement accuracy. The calculation of the absolute phase depends on the calculation of the wrapped phase and encoding technology. In this paper, a technique of obtaining the absolute phase of multi-frequency heterodyne fringe images using the Hilbert transform is presented. Since the wrapped phase can be calculated from only one fringe image of each frequency, the method does not need phase-shifting. The absolute phase can be obtained from the wrapped phase by applying the heterodyne method. The measurement time and computational complexity are dramatically reduced, the measurement efficiency is greatly improved, and this benefit from the number of images is greatly reduced. The experimental results show that the method presented in this paper performs well in the application, and the accuracy is no different from that of the phase-shifting method while the efficiency is greatly improved. [ABSTRACT FROM AUTHOR]

Simulating the unitary dynamics of a quantum system is a fundamental problem of quantum mechanics, in which quantum computers are believed to have significant advantage over their classical counterparts. One prominent such instance is the simulation of electronic dynamics, which plays an essential role in chemical reactions, non-equilibrium dynamics, and material design. These systems are time-dependent, which requires that the corresponding simulation algorithm can be successfully concatenated with itself over different time intervals to reproduce the overall coherent quantum dynamics of the system. In this paper, we quantify such simulation algorithms by the property of being fully-coherent: the algorithm succeeds with arbitrarily high success probability 1 − δ while only requiring a single copy of the initial state. We subsequently develop fully-coherent simulation algorithms based on quantum signal processing (QSP), including a novel algorithm that circumvents the use of amplitude amplification while also achieving a query complexity additive in time t, ln(1/δ), and ln(1/ϵ) for error tolerance ϵ: Θ ‖ H ‖ | t | + ln (1 / ϵ) + ln (1 / δ) . Furthermore, we numerically analyze these algorithms by applying them to the simulation of the spin dynamics of the Heisenberg model and the correlated electronic dynamics of an H2 molecule. Since any electronic Hamiltonian can be mapped to a spin Hamiltonian, our algorithm can efficiently simulate time-dependent ab initio electronic dynamics in the circuit model of quantum computation. Accordingly, it is also our hope that the present work serves as a bridge between QSP-based quantum algorithms and chemical dynamics, stimulating a cross-fertilization between these exciting fields. [ABSTRACT FROM AUTHOR]

Most models of quantum neural networks are optimized based on gradient descent, and like classical neural networks, gradient descent suffers from the barren plateau phenomenon, which reduces the effectiveness of optimization. Therefore, this paper establishes a new QNN model, the optimization process adopts efficient quantum particle swarm optimization, and tentatively adds a quantum activation circuit to our QNN model. Our model will inherit the superposition property of quantum and the random search property of quantum particle swarm. Simulation experiments on some classification data show that the model proposed in this paper has higher classification performance than the gradient descent-based QNN. [ABSTRACT FROM AUTHOR]

Steganography is widely recognized as an effective method for protecting information via digital media. This paper presents an innovative image steganography algorithm incorporating image compression, chaotic maps, and the least significant bit. The process begins with the compression of a confidential medical image using Huffman encoding. The compressed image then undergoes shuffling, facilitated by the chaotic logistic map. The bits from the shuffled image are discreetly embedded into randomly selected pixels of the cover image, guided by the chaotic piecewise smooth map. The resulting stego image is generated. Statistical analyses are applied to both the cover and stego images for evaluation. The proposed algorithm is compared against state-of-the-art algorithms, and the results demonstrate its superiority over existing methods. [ABSTRACT FROM AUTHOR]

This paper presents a novel approach—an efficient closed-form approximation iterative fitting algorithm based on laser absorption spectra. Through this closed-form approximation iterative fitting, key parameters such as peak value, spectral line width, and normalized signal area serve as indicators for iteration completion, improving the speed without compromising accuracy. Furthermore, it employs the spectral signal of n cycles as a window for further processing, minimizing external interference. The results show that the proposed method averages 9.75 iterations, while the Levenberg–Marquardt fitting method averages 60.17 iterations. The average iteration time for the proposed method is 588.83 ms, a substantial 81.7% reduction compared to the 3210.5 ms required by the Levenberg–Marquardt fitting. These results decisively demonstrate the efficacy of the proposed method in reducing iteration time and enhancing measurement precision. [ABSTRACT FROM AUTHOR]

After the service provider temporarily selects the required edge nodes based on social and storage capabilities, application execution causes the edge nodes to cache part of the application data. Therefore, offloading part of the application computing tasks to the selected edge nodes can effectively improve application execution performance. However, in cases where the resources of user's IoT devices are insufficient, tasks can be further offloaded to traditional edge servers or even to the cloud to maximize application execution efficiency. In this paper, the entire uninstall utility is modeled as a weighted sum of task completion time and energy consumption. Under the premise of considering users' preferences for completion time and energy consumption, a game-based uninstallation algorithm is proposed. The algorithm performs uninstallation by optimizing the uninstallation decision. Based on user preferences, the total system overhead is relatively small. The subsequent simulation experiments show that the algorithm can reduce system overhead on the basis of satisfying user preferences and has relatively good adaptability. [ABSTRACT FROM AUTHOR]

The application of spin–orbit torque (SOT) devices to neuromorphic computing platforms is focused on the development of hardware circuit architectures. However, the inter-device variability, the integration modes of devices and peripheral circuits, and appropriate application scenarios are still unclear, limiting the development of SOT devices in neuromorphic computing. To solve this problem, this paper first proposes a circuit compensation scheme for the difference in resistance values of SOT devices, which solves this variability problem at the circuit level. Moreover, a synergistic scheme with the circuit is developed based on the correspondence between the multistate resistance characteristics of the SOT devices and a convolutional algorithm. To achieve this, a multichannel SOT convolutional kernel circuit architecture is built, which implements an image edge recognition application. Finally, based on a simulation model, an image edge recognition hardware circuit based on our CoPt-SOT devices is implemented, which is capable of performing image edge recognition with an accuracy of 96.33%. This scheme provides technical support and development prospects for SOT devices in neural network hardware applications. [ABSTRACT FROM AUTHOR]

This paper describes the application of a machine learning (ML) algorithm using a convolution neural network, first developed in Boyer et al. ["Classification and prediction of detachment in DIII-D using neural networks trained on C III imaging," Nucl. Fusion (submitted) (2024)], to detect divertor detachment in DIII-D. Detachment detection is based on images from tangentially viewing upper and lower filtered divertor cameras that measure CIII emission at 465 nm. Separate ML models are developed for lower single null and upper single null configurations with mostly closed divertor shapes. Due to the viewing angle and divertor geometry, camera images of the upper divertor show a stark contrast in CIII emission between attached and detached conditions and the model identified detachment with 100% accuracy in the test dataset. For the lower divertor images, the contrast between attached and detached conditions is lower and the model identifies detachment with 96% accuracy. This ML model will be applied to the image data after each shot to provide a rapid assessment of divertor detachment to aid operation of DIII-D with the potential extension to other devices in the future. [ABSTRACT FROM AUTHOR]

Dispersion and multipath effects contribute to the complexity of the shallow water acoustic field. However, this complexity contains valuable information regarding both the waveguide and the acoustic source. The horizontal wavenumber and relative amplitude of the modes comprising the acoustic field are crucial pieces of information for addressing acoustic inversion problems in shallow water. However, when employing a horizontal array to extract this information, limitations arise due to array aperture and signal-to-noise ratio constraints. To attempt to solve these challenges, the approach of spatial domain dedispersion transform and frequency domain accumulation is proposed. The objective can be attained by leveraging broadband source with slowly varying phase spectrum or known phase spectrum under the constraints of small aperture arrays and low signal-to-noise ratio. Additionally, the approach is validated on dual-hydrophone horizontal array by relaxing the signal-to-noise ratio requirement. In this paper, theoretical proof of the algorithms' performance is provided, accompanied by analysis of the impact of parameters such as acoustic source bandwidth, the number of elements and array aperture. The effectiveness of the algorithms are validated through simulations and experimental data. [ABSTRACT FROM AUTHOR]

Elementary cellular automata are the simplest form of cellular automata, studied extensively by Wolfram in the 1980s. He discovered complex behavior in some of these automata and developed a classification for all cellular automata based on their phenomenology. In this paper, we present an algorithm to classify them more effectively by measuring difference patterns using the Hamming distance. Our classification aligns with Wolfram's and further categorizes them into additional subclasses. Finally, we have found a heuristic reasoning providing and explanation about why some rules evolve into fractal patterns. [ABSTRACT FROM AUTHOR]

In the last decades, numerical simulation has become a precious tool to assist the design and study of electron sources based on regular arrays of field emitters. Simulations of field emitter arrays (FEAs) require 3D treatment to account for the interactions between neighbor emitters. Therefore, modeling the thermal evolution of FEAs involves high computational resources due to the multi-physics approach and time dependency. The present paper proposes an algorithm which gives a fast prediction of the self-heating of a large array of N axisymmetric field emitters. It consists in finding for each emitter the equivalent 2D axisymmetric situation yielding the same electron current at 300 K as in the 3D array. The 3D modeling is thus efficiently split into N simulations in 2D, with a significant computation time reduction by at least one order of magnitude. The proof of concept uses 3 × 3 arrays of ideal emitters. Our results show a correct prediction, within a few percent, of the array thermo-field current and maximum temperature—two quantities of high interest for thermal failure and breakdown voltage considerations. The algorithm paves the way for including thermal effects in future optimization studies of realistic FEAs. [ABSTRACT FROM AUTHOR]

Eddy current thermography (ECT) is a non-invasive testing method that combines electromagnetic induction and infrared thermography to identify flaws in materials that conduct electricity. However, ECT faces difficulties in accurately locating and classifying defects owing to its low signal-to-noise ratio and complex defect patterns. In this paper, we propose a new method that integrates an improved faster region-convolutional neural network (R-CNN) and an adaptive Canny algorithm to enhance the defect detection performance of ECT. An improved faster R-CNN is a deep neural network that can automatically detect and locate multiple defects in a single ECT image, whereas the adaptive Canny algorithm is an edge detection technique that can identify defect boundaries. The proposed method was tested using a dataset of ECT images with different types of defects. The results demonstrated that our method achieved better accuracy, precision, and speed than existing methods. [ABSTRACT FROM AUTHOR]

We obtain the phase diagram of the hard core lattice gas with third nearest neighbor exclusion on the triangular lattice using Monte Carlo simulations that are based on a rejection-free flat histogram algorithm. In a recent paper [Darjani et al., J. Chem. Phys. 151, 104702 (2019)], it was claimed that the lattice gas with third nearest neighbor exclusion undergoes two phase transitions with increasing density with the phase at intermediate densities exhibiting hexatic order with continuously varying exponents. Although a hexatic phase is expected when the exclusion range is large, it has not been seen earlier in hard core lattice gases with short range exclusion. In this paper, by numerically determining the entropies for all densities, we show that there is only a single phase transition in the system between a low-density fluid phase and a high density ordered sublattice phase and that a hexatic phase is absent. The transition is shown to be first order in nature, and the critical parameters are determined accurately. [ABSTRACT FROM AUTHOR]

It is a challenge to simulate the hydrodynamic problems covering the large deformation of the free surface arising in severe circumstances with intense flow. This paper investigates algorithms based on the moving particle semi-implicit method for simulating large deformation and impact loads. The algorithm discretizes the fluid domain into a series of particles, each representing a part of the fluid. The pressure field calculation is implicit, and the velocity field calculation is explicit. Three models, including the gradient model, source term, and free-surface detection, have been improved and compared to determine which improvement is the best to enhance the accuracy and stability. The enhanced pressure gradient guarantees that momentum conservation can be satisfied. Particle density and velocity divergence are incompressible conditions combined in the mixed source term approach. The arc approach is used in the free-surface judging process. The results show that the combination of three models is the most effective in exploring the problems of hydrodynamic pressure and dam break. The issue of liquid sloshing including roll and sway investigates the effect of the initial distance and time step. It is found that the simulation accuracy of impact pressure can be increased as the initial distance and the time step decrease. Finally, the free surface breaking and liquid splashing phenomena are easily observed, and the method can accurately simulate the massive deformation of the free surface. These findings are helpful for hazard assessments of the various fluid mechanics-related problems. [ABSTRACT FROM AUTHOR]

Substation electrical equipment generates a massive number of infrared images during operation. However, the overall quality of the infrared images is low and it lacks image detail information. When using traditional target detection algorithms for detection, feature extraction poses great difficulties. Therefore, to address this problem, this paper proposes a target detection algorithm based on the improved faster region-based convolutional neural network (Faster R-CNN). It achieves the correct identification of different types of electrical equipment in infrared images. First, the algorithm improves the backbone network of Faster R-CNN for feature extraction. An InResNet structure is proposed to replace the residual block structure of the original ResNet-34 network, which enhances the richness of feature extraction. Second, the rectified linear unit activation function in the original feature extraction network is replaced by the exponential linear unit activation function, and group normalization is used instead of batch normalization as the network normalization method. Then, the dense connection structure is introduced into the ResNet-34 network, and the whole network is called residual dense connection network. Finally, the improved Faster R-CNN is compared to the original Faster R-CNN, a single-shot multibox detector, and you only look once v3 plus spatial pyramid pooling. The experimental results show that the improved algorithm has the highest mean average precision and average recall for most of the substation electrical equipment in infrared images. Moreover, from the confidence level of the detected electrical equipment and the accuracy of the prediction box, the improved Faster R-CNN has the best performance. [ABSTRACT FROM AUTHOR]

Link prediction has been widely studied as an important research direction. Higher-order link prediction has gained, in particular, significant attention since higher-order networks provide a more accurate description of real-world complex systems. However, higher-order networks contain more complex information than traditional pairwise networks, making the prediction of higher-order links a formidable challenging task. Recently, researchers have discovered that local features have advantages over long-range features in higher-order link prediction. Therefore, it is necessary to develop more efficient and concise higher-order link prediction algorithms based on local features. In this paper, we proposed two similarity metrics via local information, simplicial decomposition weight and closed ratio weight, to predict possible future higher-order interactions (simplices) in simplicial networks. These two algorithms capture local higher-order information at two aspects: simplex decomposition and cliques' state (closed or open). We tested their performance in eight empirical simplicial networks, and the results show that our proposed metrics outperform other benchmarks in predicting third-order and fourth-order interactions (simplices) in most cases. In addition, we explore the robustness of the proposed algorithms, and the results suggest that the performance of these novel algorithms is advanced under different sizes of training sets. [ABSTRACT FROM AUTHOR]

We report the development of a new Laplace MP2 (second-order Møller–Plesset) implementation using a range separated Coulomb potential, partitioned into short- and long-range parts. The implementation heavily relies on the use of sparse matrix algebra, density fitting techniques for the short-range Coulomb interactions, while a Fourier transformation in spherical coordinates is used for the long-range part of the potential. Localized molecular orbitals are employed for the occupied space, whereas orbital specific virtual orbitals associated with localized molecular orbitals are obtained from the exchange matrix associated with specific localized occupied orbitals. The range separated potential is crucial to achieve efficient treatment of the direct term in the MP2, while extensive screening is employed to reduce the expense of the exchange contribution in MP2. The focus of this paper is on controllable accuracy and linear scaling of the data entering the algorithm. [ABSTRACT FROM AUTHOR]

The Ramsey sequence is a canonical example of a quantum phase measurement for a spin qubit. In Ramsey measurements, the measurement efficiency can be optimized through careful selection of settings for the phase accumulation time setting, τ. This paper implements a sequential Bayesian experiment design protocol in low-fidelity Ramsey measurements, and its performance is compared to a previously reported adaptive heuristic protocol, a quantum phase estimation algorithm, and random setting choices. A workflow allowing measurements and design calculations to run concurrently largely eliminates computation time from measurement overhead. When precession frequency is the lone parameter to estimate, the Bayesian design is faster by factors of roughly 2, 4, and 5 relative to the adaptive heuristic, random τ choices, and the quantum phase estimation algorithm, respectively. When four parameters are to be determined, Bayesian experiment design and random τ choices can converge to roughly equivalent sensitivity, but the Bayesian method converges four times faster. [ABSTRACT FROM AUTHOR]

In a recent paper, Lucco Castello et al. (arXiv:2107.03537) provided an accurate parameterization of classical one-component plasma bridge functions that was embedded in a novel dielectric scheme for strongly coupled electron liquids. Here, this approach is rigorously formulated, its set of equations is formally derived, and its numerical algorithm is scrutinized. A systematic comparison with available and new path integral Monte Carlo simulations reveals a rather unprecedented agreement especially in terms of the interaction energy and the long wavelength limit of the static local field correction. [ABSTRACT FROM AUTHOR]

The feature graph (Q-H graph) is the best way to intuitively and quantitatively reflect all features of the ventilation network. In this paper, an optimized adaptive genetic algorithm is proposed to solve the problem that rectangular blocks are cut in the process of drawing a Q-H diagram of a three-dimensional ventilation network. The algorithm adopts binary coding based on node sorting and mixed genetic coding based on integer coding. The formulas for calculating the adaptive crossover rate and mutation rate are designed, which can effectively generate new individuals and get rid of the search for local optimal values, ensuring the global optimal solution. Matlab was used to test the optimization effect of the Q-H graph; the results show that, for a complex ventilation network, the improved adaptive genetic algorithm can make the Q-H graph significantly reduce the number of rectangular pieces, which is divided, and make the Q-H graph have better effect to draw clearly and intuitively. [ABSTRACT FROM AUTHOR]

Accurate and explainable artificial-intelligence (AI) models are promising tools for accelerating the discovery of new materials. Recently, symbolic regression has become an increasingly popular tool for explainable AI because it yields models that are relatively simple analytical descriptions of target properties. Due to its deterministic nature, the sure-independence screening and sparsifying operator (SISSO) method is a particularly promising approach for this application. Here, we describe the new advancements of the SISSO algorithm, as implemented into SISSO++, a C++ code with Python bindings. We introduce a new representation of the mathematical expressions found by SISSO. This is a first step toward introducing "grammar" rules into the feature creation step. Importantly, by introducing a controlled nonlinear optimization to the feature creation step, we expand the range of possible descriptors found by the methodology. Finally, we introduce refinements to the solver algorithms for both regression and classification, which drastically increase the reliability and efficiency of SISSO. For all these improvements to the basic SISSO algorithm, we not only illustrate their potential impact but also fully detail how they operate both mathematically and computationally. [ABSTRACT FROM AUTHOR]

The expensive cost of computing exact exchange in periodic systems limits the application range of density functional theory with hybrid functionals. To reduce the computational cost of exact change, we present a range-separated algorithm to compute electron repulsion integrals for Gaussian-type crystal basis. The algorithm splits the full-range Coulomb interactions into short-range and long-range parts, which are, respectively, computed in real and reciprocal space. This approach significantly reduces the overall computational cost, as integrals can be efficiently computed in both regions. The algorithm can efficiently handle large numbers of k points with limited central processing unit (CPU) and memory resources. As a demonstration, we performed an all-electron k-point Hartree–Fock calculation for LiH crystal with one million Gaussian basis functions, which was completed on a desktop computer in 1400 CPU hours. [ABSTRACT FROM AUTHOR]

In a recent paper, we proposed the adaptive shift method for correcting undersampling bias of the initiator-full configuration interaction (FCI) quantum Monte Carlo. The method allows faster convergence with the number of walkers to the FCI limit than the normal initiator method, particularly for large systems. However, in its application to some systems, mostly strongly correlated molecules, the method is prone to overshooting the FCI energy at intermediate walker numbers, with convergence to the FCI limit from below. In this paper, we present a solution to the overshooting problem in such systems, as well as further accelerating convergence to the FCI energy. This is achieved by offsetting the reference energy to a value typically below the Hartree–Fock energy but above the exact energy. This offsetting procedure does not change the exactness property of the algorithm, namely, convergence to the exact FCI solution in the large-walker limit, but at its optimal value, it greatly accelerates convergence. There is no overhead cost associated with this offsetting procedure and is therefore a pure and substantial computational gain. We illustrate the behavior of this offset adaptive shift method by applying it to the N2 molecule, the ozone molecule at three different geometries (an equilibrium open minimum, a hypothetical ring minimum, and a transition state) in three basis sets (cc-pVXZ, X = D, T, Q), and the chromium dimer in the cc-pVDZ basis set, correlating 28 electrons in 76 orbitals. We show that in most cases, the offset adaptive shift method converges much faster than both the normal initiator method and the original adaptive shift method. [ABSTRACT FROM AUTHOR]

The moth-flame algorithm shows some shortcomings in solving the complex problem of optimization, such as insufficient population diversity and unbalanced search ability. In this paper, an IMFO (Improved Moth-Flame Optimization) algorithm is proposed to be applied in solving the optimization problem of function. First, cat chaotic mapping is used to generate the initial position of moth to improve the population diversity. Second, cosine inertia weight is introduced to balance the global and local search abilities of the algorithm. Third, the memory information in the particle swarm algorithm is introduced into the iterative process of the algorithm to speed up the convergence of the population. Finally, Gaussian mutation strategy is used in the current optimal solution to avoid the algorithm from falling into the local optimum. Simulation experiments are conducted on 11 benchmark test functions, compared with other improved MFO (Moth-Flame Optimization) algorithms and classical optimization algorithms. The results show that the IMFO has higher accuracy and stability in solving the above-mentioned test functions. The proposed algorithm is experimented and verified by optimizing the KELM (Kernel Extreme Learning Machine) in an engineering example and exhibits a better optimization performance. [ABSTRACT FROM AUTHOR]

Direction-of-arrival (DOA) estimation algorithms are crucial in localizing acoustic sources. Traditional localization methods rely on block-level processing to extract the directional information from multiple measurements processed together. However, these methods assume that DOA remains constant throughout the block, which may not be true in practical scenarios. Also, the performance of localization methods is limited when the true parameters do not lie on the parameter search grid. In this paper, two trajectory models are proposed, namely the polynomial and harmonic trajectory models, to capture the DOA dynamics. To estimate trajectory parameters, two gridless algorithms are adopted: (i) Sliding Frank–Wolfe (SFW), which solves the Beurling LASSO problem, and (ii) Newtonized orthogonal matching pursuit (NOMP), which is improved over orthogonal matching pursuit (OMP) using cyclic refinement. Furthermore, our analysis is extended to include multi-frequency processing. The proposed models and algorithms are validated using both simulated and real-world data. The results indicate that the proposed trajectory localization algorithms exhibit improved performance compared to grid-based methods in terms of resolution, robustness to noise, and computational efficiency. [ABSTRACT FROM AUTHOR]

The battery needs to provide a large amount of power in a short time under the condition of a high pulse rate. Real time and accurate State of Charge (SOC) prediction can help the battery management system understand the current status of the battery better, optimize the battery charging and discharging strategy, and improve the efficiency of the battery. In order to prolong battery life and enhance battery safety, a real-time prediction algorithm for SOC of the power battery under a high pulse rate was proposed. The second order RC equivalent circuit is used to establish the model of the battery. The equivalent circuit model of the battery is designed online using the recursive least squares algorithm, and the time-varying parameter model of the battery is established. Its output value is used as the input to the gating recurrent cell neural network, and the neural network is used to output the predicted SOC value. The SOC prediction result is used as the observation vector of the adaptive extended Kalman filter algorithm to obtain the final real-time prediction result of lithium ion power battery SOC. The experimental results show that the parameters identified by the research algorithm for lithium-ion power batteries are as follows: the fluctuation range of ohmic internal resistance is 0.05–0.40 Ω, and the fluctuation range of electrochemical polarization is 0–4.5 F. The terminal voltage values collected by the research algorithm have higher accuracy, with the error being always less than 0.03 V. Moreover, the algorithm can effectively predict the SOC of lithium-ion power batteries in real time, with a maximum average absolute error of about 2%. [ABSTRACT FROM AUTHOR]

The standard fewest-switches surface hopping (FSSH) approach fails to model nonadiabatic dynamics when the electronic Hamiltonian is complex-valued and there are multiple nuclear dimensions; FSSH does not include geometric magnetic effects and does not have access to a gauge independent direction for momentum rescaling. In this paper, for the case of a Hamiltonian with two electronic states, we propose an extension of Tully's FSSH algorithm, which includes geometric magnetic forces and, through diabatization, establishes a well-defined rescaling direction. When combined with a decoherence correction, our new algorithm shows satisfying results for a model set of two-dimensional single avoided crossings. [ABSTRACT FROM AUTHOR]

We develop a Floquet surface hopping approach to deal with nonadiabatic dynamics of molecules near metal surfaces subjected to time-periodic drivings from strong light–matter interactions. The method is based on a Floquet classical master equation (FCME) derived from a Floquet quantum master equation (FQME), followed by a Wigner transformation to treat nuclear motion classically. We then propose different trajectory surface hopping algorithms to solve the FCME. We find that a Floquet averaged surface hopping with electron density (FaSH-density) algorithm works the best as benchmarked with the FQME, capturing both the fast oscillations due to the driving and the correct steady-state observables. This method will be very useful to study strong light–matter interactions with a manifold of electronic states. [ABSTRACT FROM AUTHOR]

The application of Machine Learning (ML) algorithms in chemical sciences, particularly computational chemistry, is a vastly emerging area of modern research. While many applications of ML techniques have already been in place to use ML based potential energies in various dynamical simulation studies, specific applications are also being successfully tested. In this work, the ML algorithms are tested to calculate the unimolecular dissociation time of benzene–hexachlorobenzene, benzene–trichlorobenzene, and benzene–monochlorobenzene complexes. Three ML algorithms, namely, Decision-Tree-Regression (DTR), Multi-Layer Perceptron, and Support Vector Regression are considered. The algorithms are trained with simulated dissociation times as functions (attributes) of complexes' intramolecular and intermolecular vibrational energies. The simulation data are used for an excitation temperature of 1500 K. Considering that the converged result is obtained with 1500 trajectories, an ML algorithm trained with 700 simulation points provides the same dissociation rate constant within statistical uncertainty as obtained from the converged 1500 trajectory result. The DTR algorithm is also used to predict 1000 K simulation results using 1500 K simulation data. [ABSTRACT FROM AUTHOR]

The establishment of a resistor network model has become a sharp edge to solve complex scientific problems. In this paper, we introduce Chebyshev polynomials to express the potential formula of the hammock resistor network and improve the general solution of the hammock resistor network. Moreover, through the change in different parameters, special potential formulas are proposed and displayed in 3D dynamic view. A fast algorithm of the calculating potential is given by using the matrix equation model, discrete cosine transform-II, and the fast matrix-vector multiplication. Finally, we show the advantages of our improved potential formula and fast algorithm by the calculation efficiency of the three methods. The modified potential formula and the presented fast algorithm provide a new tool for the field of science and engineering. [ABSTRACT FROM AUTHOR]

In practical application environments, objects are rarely stationary, which makes it difficult to image dynamic objects with conventional single-pixel imaging (SI) techniques. In this paper, a rotationally synchronized single-pixel imaging (RS-SI) method is proposed to image a dynamic object in rotation. The modulation pattern rotates in sync with the rapidly rotating object, and the center of object rotation is ensured to be in line with the center of the illumination pattern. Then, RS-SI reconstruction algorithm is used to reconstruct the image of the rotating object by collecting the light reflected from the object's surface. This method does not require advanced knowledge of the object rotation speed to complete the imaging. Simulation and experimental results confirm that the RS-SI can reconstruct images of an object at rotational speeds up to 422.0 rpm and can also complete imaging of a variable-speed rotating object. Imaging results of 128 × 128 pixels at a sampling rate of 27.47% can be obtained with high fidelity. The proposed RS-SI is a method for imaging an object in rotation, providing insights for future applications of single-pixel imaging technology for defect detection in rotating parts. [ABSTRACT FROM AUTHOR]

Due to factors such as meteorology and geography, the generated power of wind turbines fluctuates frequently. In this way, power changes should be predicted in grid connection to take control measures in time. In this paper, an operation and maintenance data-driven LSTM-SA (long short-term memory with self-attention) prediction algorithm is designed to predict the ultra-short-term power of wind turbines. First, the wind turbine operation and maintenance data, including wind speed, blade deflection angle, yaw angle, humidity, and temperature, are subjected to feature selection by using the Pearson correlation coefficient method and the Lasso algorithm, thereby establishing the correlation between wind speed, blade deflection angle, and out power. Then, full-connect neural network is trained to establish a mapping model of wind speed, blade deflection angle, and out power. The power change rate k is calculated by the derivative of output power to wind speed. Finally, based on the historical power data and the power change rate k, the LSTM neural network power prediction model is trained to calculate the output power prediction value. In order to increase the training efficiency and reduce the delay, the self-attention mechanism is used to optimize the hidden layer of the LSTM model. The test results show that, compared with similar prediction algorithms, this algorithm has higher prediction accuracy, faster convergence speed, and better stability, which can solve the problem of accurately predicting ultra-short-term power when wind power training data is inadequate. [ABSTRACT FROM AUTHOR]

An improved version of the mayfly algorithm called the golden annealing crossover-mutation mayfly algorithm (GSASMA) is proposed to address the low convergence efficiency and insufficient search capability of existing mayfly algorithms. First, the speed of individual mayflies is optimized using a simulated annealing algorithm to improve the update rate. The position of individuals is improved using the golden sine algorithm. Second, the impact of using different crossover and mutation methods in the algorithm is compared, and the optimal strategy is selected from the algorithm. To evaluate the performance of the algorithm, simulation experiments were carried out for 10 different test functions, and the results were compared with those of existing algorithms. The simulation results show that the algorithm developed in this paper converges faster and the solutions obtained are closer to the global optimum. Finally, GSASMA was used to optimize a support vector machine (SVM) that was used to identify the P300 signal for five subjects. The experimental results show that the SVM optimized by the algorithm proposed in this paper has higher recognition accuracy than an extreme learning machine. [ABSTRACT FROM AUTHOR]

Motor base screw loosening is a common problem in motor operation, which, if not dealt with in time, may lead to motor failure and damage. However, few studies have focused on the diagnosis and warning of this problem. Based on wavelet packet and neural network analysis, this paper presents a new algorithm for monitoring, diagnosing, and warning vibration caused by loose screws in the motor base. The vibration signal generated by the base screw loosening is monitored and sampled with sensors, and the wavelet packet is used to decompose, reconstruct, and reduce the noise of the vibration signal to enhance the time–frequency characteristics of the signal. After analyzing the fault data by wavelet, the feature vector characterizing the fault is extracted, and then, the vector and the corresponding fault type are used as the input and output of the neural network, respectively, and the non-mapping relationship between them is built to complete the diagnosis and early warning of the fault. Finally, the method is used to compare the motor base screw loosening operation and normal operation. The experiments show that the new algorithm based on wavelet packet and neural network can complete the health diagnosis and early warning of motor in the early stage of motor base screw loosening, reduce the loss caused by subsequent faults, and provide a new reference scheme for the motor fault diagnosis field. [ABSTRACT FROM AUTHOR]

An adaptive variable step-size algorithm is proposed in this paper to address the impact of the real-time acoustic feedback and the real-time secondary path identification on the overall noise reduction performance of an active noise control system. An automated adjustment weight factor is introduced in the algorithm to accelerate the convergence of the acoustic feedback path as well as the secondary path identification, and to prevent possible system divergence. It is shown in this study that the proposed algorithm can resolve the trade-off between a fast convergence and a low misalignment of the virtual and the actual control paths typically found in conventional algorithms. An optimized control structure is also proposed in the study by enabling an adaptive gain adjustment based on the output of the auxiliary filter to enhance the practicality of the control system. The effectiveness of the algorithm is tested using two simulated multi-component signals and a broadband noise signal, and the results confirm that the proposed algorithm can achieve a good noise reduction with only a few iterations. [ABSTRACT FROM AUTHOR]

Recent studies on diffusion adaptation for distributed active noise control (DANC) systems have attracted significant research interest due to their balance between computational burden and stability compared to conventional centralized and decentralized adaptation schemes. The conventional multitask diffusion FxLMS algorithm assumes that the converged solutions of all control filters are consistent to each other, which is unrealistic in practice hence results in inferior performance in noise reduction. An augmented diffusion FxLMS algorithm has been proposed to overcome this problem, which adopts a neighborhood-wide adaptation and node-based combination approach to mitigate the bias in the converged solution of the multitask diffusion algorithms. However, the improvement comes at the expense of a higher computational burden and communication cost. All existing DANC systems, including the multitask and augmented diffusion algorithms, assume one-way communication between nodes. By contrast, this paper proposes a bidirectional communication scheme for the augmented diffusion algorithm to further reduce the memory requirement, computational burden, and communication cost. Simulation results in the free field and with measured room impulse responses both demonstrate that the proposed augmented diffusion algorithm with bidirectional communication can achieve a faster convergence speed than that based on one-way communication with a lower memory, computation, and communication burden. [ABSTRACT FROM AUTHOR]

In an active headrest system virtual sensing tends to transfer the spatial zone of quiet from the residual error microphone to the ear canal. An attempt has been made in this paper to develop an auxiliary filter based virtual sensing scheme integrated with the filtered-x least mean square/fourth algorithm for an active headrest. The performance of the proposed method has been evaluated experimentally using periodic and band limited white noise. Improved noise control performance has been observed for both periodic and broadband noise. The effect of causality constraint of the performance of the algorithm has also been tested. [ABSTRACT FROM AUTHOR]

In the process of microassembly, aligning the end effectors with the micro-parts using image information is the basis of automated assembly. In order to realize the flexible and accurate clamping center locating of the micro-gripper with various shapes of jaws, this paper proposes an iterative-based processing algorithm. First, the locating problem is transformed into a multi-parameter optimization problem through the geometric analysis of the clamping process. Second, an iterative optimal algorithm based on the block coordinate descent is developed, in which a scaling golden section (SGS) scheme is proposed to calculate the iteration scaling parameters. Third, the lookup table and variable threshold iteration techniques are utilized to further improve the performance of the SGS scheme. Simulation results show that the proposed algorithm can efficiently locate the clamping center for various types of jaws with sub-pixel accuracy. Finally, a microassembly experiment is carried out to demonstrate the effectiveness of the algorithm. [ABSTRACT FROM AUTHOR]

We introduce Langevin sampling algorithms to field-theoretic simulations (FTSs) of polymers that, for the same accuracy, are ∼10× more efficient than a previously used Brownian dynamics algorithm that used predictor corrector for such simulations, over 10× more efficient than the smart Monte Carlo (SMC) algorithm, and typically over 1000× more efficient than a simple Monte Carlo (MC) algorithm. These algorithms are known as the Leimkuhler–Matthews (the BAOAB-limited) method and the BAOAB method. Furthermore, the FTS allows for an improved MC algorithm based on the Ornstein–Uhlenbeck process (OU MC), which is 2× more efficient than SMC. The system-size dependence of the efficiency for the sampling algorithms is presented, and it is shown that the aforementioned MC algorithms do not scale well with system sizes. Hence, for larger sizes, the efficiency difference between the Langevin and MC algorithms is even greater, although, for SMC and OU MC, the scaling is less unfavorable than for the simple MC. [ABSTRACT FROM AUTHOR]

We present an implementation for the calculation of molecular response properties using the algebraic-diagrammatic construction (ADC)/intermediate state representation approach. For the second-order ADC model [ADC(2)], a memory-efficient ansatz avoiding the storage of double excitation amplitudes is investigated. We compare the performance of different numerical algorithms for the solution of the underlying response equations for ADC(2) and show that our approach also strongly improves the convergence behavior for the investigated algorithms compared with the standard implementation. All routines are implemented in an open-source Python library. [ABSTRACT FROM AUTHOR]

Monte Carlo statistical ray-tracing methods are commonly employed to simulate carrier transport in nanostructured materials. In the case of a large degree of nanostructuring and under linear response (small driving fields), these simulations tend to be computationally overly expensive due to the difficulty in gathering the required flux statistics. Here, we present a novel Monte Carlo ray-tracing algorithm with computational efficiency of at least an order of magnitude compared to existing algorithms. Our new method, which is a hybrid of the analytical Boltzmann transport equation and Monte Carlo used a reduced number of ray-tracing particles, avoids current statistical challenges, such as the subtraction of two opposite going fluxes, the application of a driving force altogether, and the large simulation time required for low-energy carriers. We demonstrate the algorithm's efficiency and power in accurate simulations in large domain nanostructures with multiple defects. We believe that the new method we present is indeed more robust and user friendly compared to common methods and can enable the efficient study of transport in nanostructured materials under low-field steady-state conditions. [ABSTRACT FROM AUTHOR]

We revisit a recent proposal to model nonadiabatic problems with a complex-valued Hamiltonian through a phase-space surface hopping (PSSH) algorithm employing a pseudo-diabatic basis. Here, we show that such a pseudo-diabatic PSSH (PD-PSSH) ansatz is consistent with a quantum-classical Liouville equation (QCLE) that can be derived following a preconditioning process, and we demonstrate that a proper PD-PSSH algorithm is able to capture some geometric magnetic effects (whereas the standard fewest switches surface hopping approach cannot capture such effects). We also find that a preconditioned QCLE can outperform the standard QCLE in certain cases, highlighting the fact that there is no unique QCLE. Finally, we also point out that one can construct a mean-field Ehrenfest algorithm using a phase-space representation similar to what is done for PSSH. These findings would appear extremely helpful as far as understanding and simulating nonadiabatic dynamics with complex-valued Hamiltonians and/or spin degeneracy. [ABSTRACT FROM AUTHOR]

The fruitful interplay of high-resolution spectroscopy and quantum chemistry has a long history, especially in the field of small, semi-rigid molecules. However, in recent years, the targets of spectroscopic studies are shifting toward flexible molecules, characterized by a large number of closely spaced energy minima, all contributing to the overall spectrum. Here, artificial intelligence comes into play since it is at the basis of powerful unsupervised techniques for the exploration of soft degrees of freedom. Integration of such algorithms with a two-stage QM/QM′ (Quantum Mechanical) exploration/refinement strategy driven by a user-friendly graphical interface is the topic of the present paper. We will address in particular: (i) the performances of different semi-empirical methods for the exploration step and (ii) the comparison between stochastic and meta-heuristic algorithms in achieving a cheap yet complete exploration of the conformational space for medium sized chromophores. As test cases, we choose three amino acids of increasing complexity, whose full conformer enumeration has been reached only very recently. Next, we show that systems in condensed phases can be treated at the same level and with the same efficiency when employing a polarizable continuum description of the solvent. Finally, the challenging issue represented by the vibrational circular dichroism spectra of some rhodium complexes with flexible ligands has been addressed, showing that our fully unsupervised approach leads to remarkable agreement with the experiment. [ABSTRACT FROM AUTHOR]

Molecular dynamics simulations require barostats to be performed at a constant pressure. The usual recipe is to employ the Berendsen barostat first, which displays a first-order volume relaxation efficient in equilibration but results in incorrect volume fluctuations, followed by a second-order or a Monte Carlo barostat for production runs. In this paper, we introduce stochastic cell rescaling, a first-order barostat that samples the correct volume fluctuations by including a suitable noise term. The algorithm is shown to report volume fluctuations compatible with the isobaric ensemble and its anisotropic variant is tested on a membrane simulation. Stochastic cell rescaling can be straightforwardly implemented in the existing codes and can be used effectively in both equilibration and production phases. [ABSTRACT FROM AUTHOR]