Hypergraph Computation

Practical real-world scenarios such as the Internet,social networks,and biological networks present the challenges of data scarcity and complex correlations,which limit the applications of artificial intelligence.The graph structure is a typical tool used to formulate such correlations,it is incapable of modeling highorder correlations among different objects in systems;thus,the graph structure cannot fully convey the intricate correlations among objects.Confronted with the aforementioned two challenges,hypergraph computation models high-order correlations among data,knowledge,and rules through hyperedges and leverages these high-order correlations to enhance the data.Additionally,hypergraph computation achieves collaborative computation using data and high-order correlations,thereby offering greater modeling flexibility.In particular,we introduce three types of hypergraph computation methods:①hypergraph structure modeling,②hypergraph semantic computing,and③efficient hypergraph computing.We then specify how to adopt hypergraph computation in practice by focusing on specific tasks such as three-dimensional(3D)object recognition,revealing that hypergraph computation can reduce the data requirement by 80%while achieving comparable performance or improve the performance by 52%given the same data,compared with a traditional data-based method.A comprehensive overview of the applications of hypergraph computation in diverse domains,such as intelligent medicine and computer vision,is also provided.Finally,we introduce an open-source deep learning library,DeepHypergraph(DHG),which can serve as a tool for the practical usage of hypergraph computation.

An improvement to computational efficiency of the drain current model for double-gate MOSFET

As a connection between the process and the circuit design, the device model is greatly desired for emerging devices, such as the double-gate MOSFET. Time efficiency is one of the most important requirements for device modeling. In this paper, an improvement to the computational efficiency of the drain current model for double-gate MOSFETs is extended, and different calculation methods are compared and discussed. The results show that the calculation speed of the improved model is substantially enhanced. A two-dimensional device simulation is performed to verify the improved model. Furthermore, the model is implemented into the HSPICE circuit simulator in Verilog-A for practical application.

Fast solution to the free return orbit's reachable domain of the manned lunar mission by deep neural network

It is important to calculate the reachable domain(RD)of the manned lunar mission to evaluate whether a lunar landing site could be reached by the spacecraft. In this paper, the RD of free return orbits is quickly evaluated and calculated via the classification and regression neural networks. An efficient databasegeneration method is developed for obtaining eight types of free return orbits and then the RD is defined by the orbit’s inclination and right ascension of ascending node(RAAN) at the perilune. A classify neural network and a regression network are trained respectively. The former is built for classifying the type of the RD, and the latter is built for calculating the inclination and RAAN of the RD. The simulation results show that two neural networks are well trained. The classification model has an accuracy of more than 99% and the mean square error of the regression model is less than 0.01°on the test set. Moreover, a serial strategy is proposed to combine the two surrogate models and a recognition tool is built to evaluate whether a lunar site could be reached. The proposed deep learning method shows the superiority in computation efficiency compared with the traditional double two-body model.

Parallel Inference for Real-Time Machine Learning Applications

Hyperparameter tuning is a key step in developing high-performing machine learning models, but searching large hyperparameter spaces requires extensive computation using standard sequential methods. This work analyzes the performance gains from parallel versus sequential hyperparameter optimization. Using scikit-learn’s Randomized SearchCV, this project tuned a Random Forest classifier for fake news detection via randomized grid search. Setting n_jobs to -1 enabled full parallelization across CPU cores. Results show the parallel implementation achieved over 5× faster CPU times and 3× faster total run times compared to sequential tuning. However, test accuracy slightly dropped from 99.26% sequentially to 99.15% with parallelism, indicating a trade-off between evaluation efficiency and model performance. Still, the significant computational gains allow more extensive hyperparameter exploration within reasonable timeframes, outweighing the small accuracy decrease. Further analysis could better quantify this trade-off across different models, tuning techniques, tasks, and hardware.

A novel virtual machine deployment algorithm with energy efficiency in cloud computing

In order to improve the energy efficiency of large-scale data centers, a virtual machine(VM) deployment algorithm called three-threshold energy saving algorithm(TESA), which is based on the linear relation between the energy consumption and(processor) resource utilization, is proposed. In TESA, according to load, hosts in data centers are divided into four classes, that is,host with light load, host with proper load, host with middle load and host with heavy load. By defining TESA, VMs on lightly loaded host or VMs on heavily loaded host are migrated to another host with proper load; VMs on properly loaded host or VMs on middling loaded host are kept constant. Then, based on the TESA, five kinds of VM selection policies(minimization of migrations policy based on TESA(MIMT), maximization of migrations policy based on TESA(MAMT), highest potential growth policy based on TESA(HPGT), lowest potential growth policy based on TESA(LPGT) and random choice policy based on TESA(RCT)) are presented, and MIMT is chosen as the representative policy through experimental comparison. Finally, five research directions are put forward on future energy management. The results of simulation indicate that, as compared with single threshold(ST) algorithm and minimization of migrations(MM) algorithm, MIMT significantly improves the energy efficiency in data centers.

Computational fluid dynamics(CFD) simulation of effect of baffles on separation in mixer settler

The main ideas in the development of the solvent extraction mixer settler focused on achieving clean phase separation,minimizing the loss of the reagents and decreasing the surface area of the settlers.The role of baffles in a mechanically agitated vessel is to ensure even distribution,reduce settler turbulence,promote the stability of power drawn by the impeller and to prevent swirling and vortexing of liquid,thus,greatly improving the mixing of liquid.The insertion of the appropriate number of baffles clearly improves the extent of liquid mixing.However,excessive baffling would interrupt liquid mixing and lengthen the mixing time.Computational fluid dynamics(CFD) provides a tool for determining detailed information on fluid flow(hydrodynamics) which is necessary for modeling subprocesses in mixer settler.A total of 54 final CFD runs were carried out representing different combinations of variables like number of baffles,density and impeller speed.CFD data shows that amount of separation increases with increasing baffles number and decreasing impeller speed.

Efficient computation for dynamic responses of systems with time-varying characteristics

Based on Neumman series and epsilon-algorithm, an efficient computation for dynamic responses of systems with arbitrary time-varying characteristics is investigated. Avoiding the calculation for the inverses of the equivalent stiffness matrices in each time step, the computation effort of the proposed method is reduced compared with the full analysis of Newmark method. The validity and applications of the proposed method are illustrated by a 4-DOF spring-mass system with periodical time-varying stiffness properties and a truss structure with arbitrary time-varying lumped mass. It shows that good approximate results can be obtained by the proposed method compared with the responses obtained by the full analysis of Newmark method.

Efficiency analysis of numerical integrations for finite element substructure in real-time hybrid simulation

Finite element（FE） is a powerful tool and has been applied by investigators to real-time hybrid simulations（RTHSs）. This study focuses on the computational efficiency, including the computational time and accuracy, of numerical integrations in solving FE numerical substructure in RTHSs. First, sparse matrix storage schemes are adopted to decrease the computational time of FE numerical substructure. In this way, the task execution time（TET） decreases such that the scale of the numerical substructure model increases. Subsequently, several commonly used explicit numerical integration algorithms, including the central difference method（CDM）, the Newmark explicit method, the Chang method and the Gui-λ method, are comprehensively compared to evaluate their computational time in solving FE numerical substructure. CDM is better than the other explicit integration algorithms when the damping matrix is diagonal, while the Gui-λ（λ = 4） method is advantageous when the damping matrix is non-diagonal. Finally, the effect of time delay on the computational accuracy of RTHSs is investigated by simulating structure-foundation systems. Simulation results show that the influences of time delay on the displacement response become obvious with the mass ratio increasing, and delay compensation methods may reduce the relative error of the displacement peak value to less than 5% even under the large time-step and large time delay.

STUDY ON EFFICIENT COMPUTATION AND PERFORMANCE OF AV-BASED REDUCED-RANK FILTERING

In this paper, the complexity and performance of the Auxiliary Vector (AV) based reduced-rank filtering are addressed. The AV filters presented in the previous papers have the general form of the sum of the signature vector of the desired signal and a set of weighted AVs,which can be classified as three categories according to the orthogonality of their AVs and the optimality of the weight coefficients of the AVs. The AV filter with orthogonal AVs and optimal weight coefficients has the best performance, but requires considerable computational complexity and suffers from the numerical unstable operation. In order to reduce its computational load while keeping the superior performance, several low complexity algorithms are proposed to efficiently calculate the AVs and their weight coefficients. The diagonal loading technique is also introduced to solve the numerical unstability problem without complexity increase. The performance of the three types of AV filters is also compared through their application to Direct Sequence Code Division Multiple Access (DS-CDM A) systems for interference suppression.

Bubble size modeling approach for the simulation of bubble columns

The constant bubble size modeling approach(CBSM)and variable bubble size modeling approach(VBSM)are frequently employed in Eulerian–Eulerian simulation of bubble columns.However,the accuracy of CBSM is limited while the computational efficiency of VBSM needs to be improved.This work aims to develop method for bubble size modeling which has high computational efficiency and accuracy in the simulation of bubble columns.The distribution of bubble sizes is represented by a series of discrete points,and the percentage of bubbles with various sizes at gas inlet is determined by the results of computational fluid dynamics(CFD)–population balance model(PBM)simulations,whereas the influence of bubble coalescence and breakup is neglected.The simulated results of a 0.15 m diameter bubble column suggest that the developed method has high computational speed and can achieve similar accuracy as CFD–PBM modeling.Furthermore,the convergence issues caused by solving population balance equations are addressed.

High-speed corner detection based on fuzzy ID3 decision tree

A high-speed comer detection algorithm based on fuzzy ID3 decision tree was proposed. In the algorithm, the Bresenham circle with 3-pixel radius was used as the test mask, overlapping the candidate comers with the nucleus. Connected pixels on the circle were applied to compare the intensity value with the nucleus, with the membership function used to give the fuzzy result. The pixel with maximum information gain was chosen as the parent node to build a binary decision tree. Thus, the comer detector was derived. The pictures taken in Fengtai Railway Station in Beijing were used to test the method. The experimental results show that when the number of pixels on the test mask is chosen to be 9, best result can be obtained. The comer detector significantly outperforms existing detector in computational efficiency without sacrificing the quality and the method also provides high performance against Poisson noise and Gaussian blur.

Multivariate adaptive regression splines and neural network models for prediction of pile drivability

Piles are long, slender structural elements used to transfer the loads from the superstructure through weak strata onto stiffer soils or rocks. For driven piles, the impact of the piling hammer induces compression and tension stresses in the piles. Hence, an important design consideration is to check that the strength of the pile is sufficient to resist the stresses caused by the impact of the pile hammer. Due to its complexity, pile drivability lacks a precise analytical solution with regard to the phenomena involved.In situations where measured data or numerical hypothetical results are available, neural networks stand out in mapping the nonlinear interactions and relationships between the system’s predictors and dependent responses. In addition, unlike most computational tools, no mathematical relationship assumption between the dependent and independent variables has to be made. Nevertheless, neural networks have been criticized for their long trial-and-error training process since the optimal configuration is not known a priori. This paper investigates the use of a fairly simple nonparametric regression algorithm known as multivariate adaptive regression splines（MARS）, as an alternative to neural networks, to approximate the relationship between the inputs and dependent response, and to mathematically interpret the relationship between the various parameters. In this paper, the Back propagation neural network（BPNN） and MARS models are developed for assessing pile drivability in relation to the prediction of the Maximum compressive stresses（MCS）, Maximum tensile stresses（MTS）, and Blow per foot（BPF）. A database of more than four thousand piles is utilized for model development and comparative performance between BPNN and MARS predictions.

Application of Improved Hybrid Interface Substructural Component Modal Synthesis Method in Vibration Characteristics of Mistuned Blisk

The large and complex structures are divided into hundreds of thousands or millions degrees of freedom（DOF） when they are calculated which will spend a lot of time and the efficiency will be extremely low. The classical component modal synthesis method （CMSM） are used extensively, but for many structures in the engineering of high-rise buildings, aerospace systemic engineerings, marine oil platforms etc, a large amount of calculation is still needed. An improved hybrid interface substructural component modal synthesis method（HISCMSM） is proposed. The parametric model of the mistuned blisk is built by the improved HISCMSM. The double coordinating conditions of the displacement and the force are introduced to ensure the computational accuracy. Compared with the overall structure finite element model method（FEMM）, the computational time is shortened by23.86%–31.56%and the modal deviation is 0.002%–0.157% which meets the requirement of the computational accuracy. It is faster 4.46%–10.57% than the classical HISCMSM. So the improved HISCMSM is better than the classical HISCMSM and the overall structure FEMM. Meanwhile, the frequency and the modal shape are researched, considering the factors including rotational speed, gas temperature and geometry size. The strong localization phenomenon of the modal shape’s the maximum displacement and the maximum stress is observed in the second frequency band and it is the most sensitive in the frequency veering. But the localization phenomenon is relatively weak in 1st and the 3d frequency band. The localization of the modal shape is more serious under the condition of the geometric dimensioning mistuned. An improved HISCMSM is proposed, the computational efficiency of the mistuned blisk can be increased observably by this method.

Efficient Mass Transport and Electrochemistry Coupling Scheme for Reliable Multiphysics Modeling of Planar Solid Oxide Fuel Cell Stack

A multiphysics model for a production scale planar solid oxide fuel cell （SOFC） stack is important for the SOFC technology, but usually requires an unpractical amount of computing resource. The major cause for the huge computing resource requirement is identified as the need to solve the cathode O2 transport and the associated electrochemistry. To overcome the technical obstacle, an analytical model for solving the O2 transport and its coupling with the electrochemistry is derived. The analytical model is used to greatly reduce the numerical mesh complexity of a multiphysics model. Numerical test shows that the analytical approximation is highly accurate and stable. A multiphysics numerical modeling tool taking advantage of the analytical solution is then developed through Fluent@. The numerical efficiency and stability of this modeling tool are further demonstrated by simulating a 30- cell stack with a production scale cell size. Detailed information about the stack performance is revealed and briefly discussed. The multiphysics modeling tool can be used to guide the stack design and select the operating parameters.

A PREDICTION TECHNIQUE FOR DYNAMIC ANALYSIS OF FLAT PLATES IN MID-FREQUENCY RANGE

The wave-based method （WBM） has been applied for the prediction of mid-frequency vibrations of fiat plates. The scaling factors, Gauss point selection rule and truncation rule are introduced to insure the wave model to converge. Numerical results show that the prediction tech- nique based on WBM is with higher accuracy and smaller computational effort than the one on FEM, which implies that this new technique on WBM can be applied to higher-frequency range.

Multi-resolution nonlinear topology optimization with enhanced computational efficiency and convergence

Huge calculation burden and difficulty in convergence are the two central conundrums of nonlinear topology optimization(NTO).To this end,a multi-resolution nonlinear topology optimization(MR-NTO)method is proposed based on the multiresolution design strategy(MRDS)and the additive hyperelasticity technique(AHT),taking into account the geometric nonlinearity and material nonlinearity.The MR-NTO strategy is established in the framework of the solid isotropic material with penalization(SIMP)method,while the Neo-Hookean hyperelastic material model characterizes the material nonlinearity.The coarse analysis grid is employed for finite element(FE)calculation,and the fine material grid is applied to describe the material configuration.To alleviate the convergence problem and reduce sensitivity calculation complexity,the software ANSYS coupled with AHT is utilized to perform the nonlinear FE calculation.A strategy for redistributing strain energy is proposed during the sensitivity analysis,i.e.,transforming the strain energy of the analysis element into that of the material element,including Neo-Hooken and second-order Yeoh material.Numerical examples highlight three distinct advantages of the proposed method,i.e.,it can(1)significantly improve the computational efficiency,(2)make up for the shortcoming that NTO based on AHT may have difficulty in convergence when solving the NTO problem,especially for 3D problems,(3)successfully cope with high-resolution 3D complex NTO problems on a personal computer.

A new spatial coherence model and analytical coeffi cients for multi-support response spectrum combination

In this paper, a new spatial coherence model of seismic ground motions is proposed by a fitting procedure. The analytical expressions of modal combination （correlation） coefficients of structural response are developed for multi-support seismic excitations. The coefficients from both the numerical integration and analytical solutions are compared to verify the accuracy of the solutions. It is shown that the analytical expressions of numerical modal combination coefficients are of high accuracy. The results of random responses of an example bridge show that the analytical modal combination coefficients developed in this paper are accurate enough to meet the requirements needed in practice. In addition, the computational efficiency of the analytical solutions of the modal combination coefficients is demonstrated by the response computation of the example bridge. It is found that the time required for the structural response analysis by using the analytical modal combination coefficients is less than 1/20 of that using numerical integral methods.

Robust entry guidance using multi-segment linear pseudospectral model predictive control

This paper presents a robust entry guidance algorithm for the high lift-to-drag ratio entry vehicle that employs the recently developed pseudospectral model predict control in a segmented manner. Here, the guidance commands are the longitudinal lift-to-drag (L/D) and bank reversal commands, which are calculated by successively solving multiple segment linear algebraic equations. These equations are derived using the linear pseudospectral method, control parametrization and calculus of variations. The method uses orthogonal polynomials and computes the updates through a series of analytical formulae, which makes it accurate and computationally effective. Moreover, it is able to adjust the number of bank reversals by providing the precise bank reversal point so as to fully exploit the potential of lateral maneuver. The method also employs proportional navigation and polynomial guidance after the last bank reversal to meet multiple terminal constraints. High-fidelity numerical simulations with various destinations are carried out to demonstrate its applicability. Furthermore, Monte Carlo simulations are also conducted to show that the proposed algorithm consistently offers very stable and robust performances and has superior performances in computational efficiency, guidance accuracy and lateral trajectory shaping capability in comparison with other typical methods. © 1990-2011 Beijing Institute of Aerospace Information.

A High-Accuracy Single Patch Representation of Multi-Patch Geometries with Applications to Isogeometric Analysis

This paper presents a novel approximating method to construct highprecision single-patch representation of B-spline surface from a multi-patch representation for isogeometric applications.In isogeometric analysis,multi-patch structure is not easy to achieve high continuity between neighboring patches which will reduce the advantage of isogeometric analysis in a sense.The proposed method can achieve high continuity at surface stitching region with low geometric error,and this technique exploits constructing the approximate surface with several control points are from original surfaces,which guarantees the local feature of the surface can be well-preserved with high precision.With the proposed approximating method,isogeometric analysis results using the new single-patch can be obtained efficiently compared with the original multi-patch structure.Several examples are presented to illustrate the effectiveness,accuracy and efficiency of the proposed method.

A meshfree-based local Galerkin method with condensation of degree of freedom for elastic dynamic analysis

Condensation technique of degree of freedom is first proposed to improve the computational efficiency of meshfree method with Galerkin weak form for elastic dy- namic analysis. In the present method, scattered nodes with- out connectivity are divided into several subsets by cells with arbitrary shape. Local discrete equation is established over each cell by using moving Kriging interpolation, in which the nodes that located in the cell are used for approxima- tion. Then local discrete equations can be simplified by con- densation of degree of freedom, which transfers equations of inner nodes to equations of boundary nodes based on cells. The global dynamic system equations are obtained by as- sembling all local discrete equations and are solved by using the standard implicit Newmark＇s time integration scheme. In the scheme of present method, the calculation of each cell is carried out by meshfree method, and local search is imple- mented in interpolation. Numerical examples show that the present method has high computational efficiency and good accuracy in solving elastic dynamic problems.