最新刊期
      卷 62 ， 期 5 ， 2023

        Column

      • SHA Xingyu,ZHANG Jiaqi,YOU Keyou
        Vol. 62, Issue 5, Pages: 1-23(2023) DOI: 10.13471/j.cnki.acta.snus.2023A023
        摘要：We study distributed optimization problems over a directed network, where nodes aim to minimize the sum of local objective functions via directed communications with neighbors. Many algorithms are designed to solve it for synchronized or randomly activated implementation, which may create deadlocks in practice. In sharp contrast, we propose a fully asynchronous push-pull gradient(APPG) algorithm, where each node updates without waiting for any other node by using possibly delayed information from neighbors. Then, we construct two novel augmented networks to analyze asynchrony and delays, and quantify its convergence rate from the worst-case point of view. Particularly, all nodes of APPG converge to the same optimal solution at a linear rate of 𝒪λk if local functions have Lipschitz-continuous gradients and their sum satisfies the Polyak-Łojasiewicz condition (convexity is not required), where λ∈0,1 is explicitly given and the virtual counter k increases by one when any node updates. Finally, the advantage of APPG over the synchronous counterpart and its linear speedup efficiency are numerically validated via a logistic regression problem.  
        关键词：fully asynchronous;distributed optimization;linear convergence;Polyak-Łojasiewicz condition   
        67
        |
        16
        |
        0
        <HTML>
        <L-PDF><Meta-XML>
        <引用本文> <批量引用> 40983328 false
        发布时间：2023-11-07
      • ZHANG Zhiyuan,ZHU Xuening
        Vol. 62, Issue 5, Pages: 24-37(2023) DOI: 10.13471/j.cnki.acta.snus.2023A027
        摘要：Network autoregression and factor model are effective methods for modeling network time series data. In this study, we propose a network autoregression model with a factor structure that incorporates a latent group structure to address nodal heterogeneity within the network. An iterative algorithm is employed to minimize a least-squares objective function, allowing for simultaneous estimation of both the parameters and the group structure. To determine the unknown number of groups and factors, a PIC criterion is introduced. Additionally, statistical inference of the estimated parameters is presented. To assess the validity of the proposed estimation and inference procedures, we conduct extensive numerical studies. We also demonstrate the utility of our model using a stock dataset obtained from the Chinese A-Share stock market.  
        关键词：Network autoregression;factor structure;heterogeneity;latent group structure;network time series   
        40
        |
        10
        |
        0
        <HTML>
        <L-PDF><Meta-XML>
        <引用本文> <批量引用> 41018200 false
        发布时间：2023-11-07
      • HUANG Yu,HUANG Longbo
        Vol. 62, Issue 5, Pages: 38-49(2023) DOI: 10.13471/j.cnki.acta.snus.2023A022
        摘要：Deep multi-modal learning, a rapidly growing field with a wide range of practical applications, aims to effectively utilize and integrate information from multiple sources, known as modalities. Despite its impressive empirical performance, the theoretical foundations of deep multi-modal learning have yet to be fully explored. In this paper, we will undertake a comprehensive survey of recent developments in multi-modal learning theories, focusing on the fundamental properties that govern this field. Our goal is to provide a thorough collection of current theoretical tools for analyzing multi-modal learning, to clarify their implications for practitioners, and to suggest future directions for the establishment of a solid theoretical foundation for deep multi-modal learning.  
        关键词：multi-modal learning;machine learning theory;optimization;generalization   
        59
        |
        8
        |
        0
        <HTML>
        <L-PDF><Meta-XML>
        <引用本文> <批量引用> 40984348 false
        发布时间：2023-11-07
      • YANG Zai,WU Xunmeng,XU Zongben
        Vol. 62, Issue 5, Pages: 50-58(2023) DOI: 10.13471/j.cnki.acta.snus.2023A040
        摘要：Spectral compressed sensing refers to the problem of recovering spectral-sparse signals and their continuous-valued frequency parameters from limited samples. It is an extension of the classical spectral analysis problem of signals and widely used in information technology fields such as array and radar signal processing and wireless communications. The classical methods for spectral compressed sensing and the convex relaxation methods of this century have limits in the scope of application， estimation accuracy or algorithm speed， which cannot satisfy the urgent demands for high accuracy and speed in technologies such as current 5G and future 6G wireless communications. Recently， a series of nonconvex optimization models based on structured low-rank matrices have been proposed. By characterizing the geometric structures of spectral sparse signals， the original highly non-convex optimization problem in the parameter domain is cast as a structured low-rank matrix recovery problem in the signal domain， which provides a novel solution for the spectral compressed sensing problem and brings a substantial improvement in the algorithm accuracy. In this paper， we systematically review the existing structured low-rank matrix recovery models and algorithms for three types of spectral sparse signals： single-channel， multichannel， and constant modulus， analyze commonalities and differences of these models； and point out possible future research directions.  
        关键词：spectral compressed sensing;spectral analysis of signals;structured low-rank matrix recovery;nonconvex optimization   
        123
        |
        10
        |
        0
        <HTML>
        <L-PDF><Meta-XML>
        <引用本文> <批量引用> 40983035 false
        发布时间：2023-11-07
      • BAO Chenglong,WEI Fuchao
        Vol. 62, Issue 5, Pages: 59-66(2023) DOI: 10.13471/j.cnki.acta.snus.2023A035
        摘要：Anderson mixing is a classical extrapolation method. It can make use of the information in historical iterations to accelerate the convergence of fixed-point iterations， and has been successfully applied in scientific computing and machine learning. Since Anderson mixing often exhibits superior numerical performance in practice， the algorithm design and theoretical analysis around Anderson mixing in various applications have become hot topics in recent years. This article reviews the research advance on Anderson mixing， and highlights new algorithms based on Anderson mixing.  
        关键词：Anderson mixing;fixed-point iteration;Krylov subspace method;quasi-Newton method   
        55
        |
        2
        |
        0
        <HTML>
        <L-PDF><Meta-XML>
        <引用本文> <批量引用> 40984332 false
        发布时间：2023-11-07

        Research articles

      • WANG Zhuolin,JIANG Junyang,YANG Gengchao,YAO Qinghe,JIANG Zichao,ZHANG Yi
        Vol. 62, Issue 5, Pages: 67-77(2023) DOI: 10.13471/j.cnki.acta.snus.2022A064
        摘要：As a meshless method， the moving particle semi-implicit（MPS） method forms the pressure Poisson equation by using the incompressibility of fluid， which not only obtain the accuracy of pressure but also bring a high cost of calculation. Therefore， it is not appropriate for MPS to solve large-scale fluid simulations. IA new algorithm NN-MPS is proposed to solve the above problems， which transforms the solution of the Poisson equation into a regression problem by using neural network. The NN-MPS algorithm realizes the quick solution of the Poisson equation by constructing the prediction model of flow field features and pressure at each step. In this work， the procedure of solving the pressure Poisson equation is further transported to Atlas 200 DK device for a faster speed of solving procedure. Results show that the acceleration method of MPS mentioned in this work has the characteristics of low cost， high speed， and low accuracy loss， and the solution speed has been improved. We also verified the feasibility of applying the edge computing device in the field of CFD.  
        关键词：edge computing;neural network;NN-MPS;CFD   
        21
        |
        3
        |
        0
        <HTML>
        <L-PDF><Meta-XML>
        <引用本文> <批量引用> 38963275 false
        发布时间：2023-11-07
      • WANG Deliang,LIU Jike,LIU Guang
        Vol. 62, Issue 5, Pages: 78-84(2023) DOI: 10.13471/j.cnki.acta.snus.2023A001
        摘要：The incremental harmonic balance method （IHB method） is a semi-numerical and semi-analytical method for strongly nonlinear dynamic systems. However， previous studies have shown that the convergence performance of the original IHB method in solving systems with multiple solutions strongly depends on the selection of initial values. The Tikhonov regularization is often used in optimization problems to solve potential ill-posed problems. In this paper， by incorporating the Tikhonov regularization into the original IHB method， an improved IHB method （TIHB method） is proposed to obtain the multiple solutions of the Mathieu-Duffing system. The results show that the improved TIHB method can obtain the stable and unstable solutions of the Mathieu-Duffing system quickly and efficiently， and the convergence performance of the TIHB method is much better than the original IHB method.  
        关键词：nonlinear vibration;IHB method;Tikhonov regularization;multiple solution   
        20
        |
        2
        |
        0
        <HTML>
        <L-PDF><Meta-XML>
        <引用本文> <批量引用> 39666776 false
        发布时间：2023-11-07
      • LU Haoyu,ZHUANG Hongcheng,CHEN Zengping,PANG Gaokun
        Vol. 62, Issue 5, Pages: 85-91(2023) DOI: 10.13471/j.cnki.acta.snus.2023B006
        摘要：With the increasing densification of wireless networks and the need for multifaced network coverage and capacity, achieving stable network optimization becomes more challenging for SON technology. Aiming at the problem of weak coverage resulting from interference and obstacle blocking in wireless network， this paper proposes a gridization based RIS-aided optimization scheme. The weak coverage problem is formulated to maximize the received signal strength of the coverage problem area and its optimization variables are the location， incident angle and phase shift of RIS. Since traditional decoupling processing struggles to find the optimal solution， the scheme proposed by this paper specifies the set of reflection angles by rasterizing the coverage problem area， and then rapidly find the optimal phase shift of RIS， which simplifies the solving of original optimization problem significantly.  
        关键词：RIS;self-organizing network;millimeter wave;coverage optimization   
        27
        |
        3
        |
        0
        <HTML>
        <L-PDF><Meta-XML>
        <引用本文> <批量引用> 39081338 false
        发布时间：2023-11-07
      • WEI Zhongshuang,HOU Wei,ZHAO Yan,ZHENG Shousen,FU Qing
        Vol. 62, Issue 5, Pages: 92-100(2023) DOI: 10.13471/j.cnki.acta.snus.2023B012
        摘要：The state of charge of the battery is one of the fundamental parameters of the lithium-ion battery， which can't be measured directly. In this paper， the second-order Thevenin model is used to analyze the characteristics of lithium-iron phosphate battery， and the model parameters at different SoC are identified through the hybrid pulse power characteristic （HPPC） experiment. Based on MATLAB/Simulink platform， an estimation system combining variable parameter second-order Thevenin model and extended Kalman filter （EKF） algorithm is built. The simulation results show that the simulation error of the estimation system is less than 2.5% under different working conditions， which provides support for the development of lithium-ion battery management system.  
        关键词：lithium-iron phosphate battery;SoC;the second-order Thevenin model;parameter identify;EKF algorithm   
        18
        |
        1
        |
        0
        <HTML>
        <L-PDF><Meta-XML>
        <引用本文> <批量引用> 38820673 false
        发布时间：2023-11-07
      • SONG Haili,HUANG Rong
        Vol. 62, Issue 5, Pages: 101-106(2023) DOI: 10.13471/j.cnki.acta.snus.2023B021
        摘要：A SrTiO3 / SrTiO3 homogeneous epitaxial film was grown along the SrTiO3 substrate （001） by PLD method. The film was conductive before annealing， and became insulating states after annealing. A variety of advanced spherical aberration-corrected transmission microscopy techniques like annular high-angle darkfield image， annular brightfield image and atomic-resolution energy loss spectroscope （EELS） was used to explore the atomic occupancy， electronic structure and oxygen vacancies in the film. A TiOx (1<x<2) reconstruction layer was found on the surface of the film. And the valence state of Ti in the reconstruction layer increases slightly after annealing. According to EELS analysis， oxygen vacancies existed in the whole film including the surface and interface before annealing， and disappeared after annealing. Therefore， for non-polar films SrTiO3/SrTiO3， oxygen vacancies near the film surface and interface maybe the origin of electrical conduction.  
        关键词：perovskite oxides;spherical aberration-corrected transmission electron microscope;energy loss spectroscope;oxygen vacancy   
        20
        |
        1
        |
        0
        <HTML>
        <L-PDF><Meta-XML>
        <引用本文> <批量引用> 39125635 false
        发布时间：2023-11-07
      • LUO Renyu,LI Qizhi,HUANG Yunjin,YANG Gengchao,YU Maomao,ZU Gongbo,YAO Qinghe
        Vol. 62, Issue 5, Pages: 107-114(2023) DOI: 10.13471/j.cnki.acta.snus.2023B020
        摘要：Based on the Guangzhou-Dongguan-Shenzhen Intercity Railway Project， we use the discrete element method to simulate the blasting process of deep rock masses in foundation pits， and analyzes the safety of the surrounding environment. Firstly， the rationality of PFC simulation blasting was verified. Subsequently， reasonable geological microscopic parameters were obtained through uniaxial compression tests； and a two-dimensional foundation pit model was established to study the impact of ground disturbance from three perspectives， referring to different initial stresses， burial depths and the presence of cracks. The results indicate that：the greater the initial stress， the greater the corresponding ground velocity； the variation curve of ground velocity with the burial depth of explosives shows an upward parabolic shape； the presence of cracks can effectively reduce the peak ground velocity by 13%．  
        关键词：building pit blasting;PFC;blasting stress wave;speed monitoring;pre-cracks   
        25
        |
        5
        |
        0
        <HTML>
        <L-PDF><Meta-XML>
        <引用本文> <批量引用> 39081376 false
        发布时间：2023-11-07
      • LIAO Yixiao,ZHENG Min,ZHANG Luxing,YU Fayuan,DUAN Yanhui
        Vol. 62, Issue 5, Pages: 115-127(2023) DOI: 10.13471/j.cnki.acta.snus.2022E047
        摘要：Distributed electric propulsion aircraft improve the lift-drag characteristics of the aircraft by using propellers distributed along the leading edge of the wing to generate slipstream. In this paper， we use a wing configuration similar to the X-57 and study the effect of propeller tilt on the wing aerodynamics using an actuator disk method. We introduce the actuator disk method and validate it with experimental data and CFD calculation data. We compare the lift-drag characteristics of the take-off and landing configurations with and without slipstream for different angles of attack. The effect of the propeller tilt angle on the aerodynamic performance of the wing is analyzed respectively for the take-off and landing configurations of the wing. The results show that propeller tilt changes the effective angle of attack and thrust direction of the slipstream on the wing. When the effective angle of attack decreases， the aerodynamic lift coefficient of the wing also decreases， but the component of propeller thrust in the lift direction due to tilt increases the effective lift coefficient of the wing. At 5.5° angle of attack， for the take-off configuration， the maximum effective lift coefficient is obtained when the propeller tilt angle is 20.18°， which is about 2.445 1， and for the landing configuration， the maximum effective lift coefficient is obtained when the propeller tilt angle is 23.83°， which is about 3.628 8.  
        关键词：distributed electric propulsion;actuator disk;propeller tilt;effective lift coefficient   
        19
        |
        4
        |
        0
        <HTML>
        <L-PDF><Meta-XML>
        <引用本文> <批量引用> 38804803 false
        发布时间：2023-11-07
      • CHEN Xinjun,LU Zhongrong,WANG Li
        Vol. 62, Issue 5, Pages: 128-135(2023) DOI: 10.13471/j.cnki.acta.snus.2022E053
        摘要：In practical engineering application， the displacement measurement usually needs to set reference points， which makes it difficult to directly measure the displacement of large structures. Therefore， this paper proposes a new method which directly uses easily accessible （no reference point needed） and highly precise acceleration data to recover the displacement of the structure. Firstly， the discrete linear equation between displacement and acceleration is established. Displacement is the double integral of acceleration by directly solving the equation. However， direct integration is very sensitive to low frequency noise， resulting in "low-frequency drift" phenomenon. Secondly， Tikhonov regularization is introduced on the basis of discrete linear equations to overcome the ill-posed problem and solve the "low-frequency drift" problem. In addition， in order to improve the accuracy of displacement reconstruction， a displacement reconstruction scheme based on overlapping windows is developed， and the interval length of each window should be about 3 times of the main frequency period of the signal. It is found that this method is suitable for the displacement reconstruction of acceleration at single main frequency （or narrow frequency）， but it is usually poor for broadband displacement reconstruction. In order to improve the accuracy of broadband displacement reconstruction， this paper further proposes a bandlimited signal decomposition method， which decomposes the broadband acceleration signal into several narrow frequency signals， and then uses the regularization method to reconstruct the displacement of each narrow frequency signal with high accuracy. Finally， numerical and experimental results show that the proposed method can accurately reconstruct the displacement of the structure from the broadband acceleration signal.  
        关键词：broadband displacement reconstruction;low-frequency drift;bandlimited signal decomposition;regularization   
        15
        |
        1
        |
        0
        <HTML>
        <L-PDF><Meta-XML>
        <引用本文> <批量引用> 38804832 false
        发布时间：2023-11-07
      • ZHANG Lan,JIANG Zichao,CHEN Yuhui,ZHANG Yi,YANG Gengchao,YAO Qinghe
        Vol. 62, Issue 5, Pages: 136-144(2023) DOI: 10.13471/j.cnki.acta.snus.2023D008
        摘要：The study of the time information and transitional behavior of flow states in the development of intermittent solutions in Rayleigh-Bénard （RB） convection of viscoelastic fluids has great significance in understanding the flow mechanics of Phan-Thien-Tanner （PTT） viscoelastic fluid RB convection. In this paper， we first use a second-order upwind-type total variation diminishing （TVD） scheme to numerically simulate PTT viscoelastic fluid RB convection， and demonstrate the convergence and second-order accuracy of the numerical method in spatial discretization based on the comparison between the numerical and analytical solutions. The numerical results indicate that the numerical method employed in this paper can simulate the evolution of PTT viscoelastic fluid RB thermal convection with discontinuity initial field， and provide a fundamental explanation for the transient flow phenomena of this flow up to the development of steady periodic solutions， which are consistent with the flow phenomena developed from the a priori velocity field.  
        关键词：discontinuous solution;total variation diminishing scheme;viscoelastic fluid;Phan-Thien-Tanner model;Rayleigh-Bénard convection   
        21
        |
        1
        |
        0
        <HTML>
        <L-PDF><Meta-XML>
        <引用本文> <批量引用> 38800107 false
        发布时间：2023-11-07
      • SONG Lang,CHAI Shaobo,SONG Boyang,CHAI Lianzeng,DU Yuxiang
        Vol. 62, Issue 5, Pages: 145-156(2023) DOI: 10.13471/j.cnki.acta.snus.2023D007
        摘要：The characteristics of filled joints are important factors affecting the stability of rock engineering. To study the influence of the characteristics of filled joints on the statically compressive mechanics and deforming features of rock， based on laboratory uniaxial tests of rock samples with filled joints， we performed particle flow simulations to study the evolutions of mechanical parameters， micro-crack development， and the final failure morphology characteristics considering the variety of angles， numbers， and thicknesses of filled joints. The results show that the peak strength and elastic modulus display a U-shaped variation with the increase of the angle of joints， monotonically decrease with the thickness of joints and the decreasing speed of elastic modulus gradually slows down. The crack initiation stress and the total number of micro-cracks decrease first and then increase with the joint angle， and the total number of micro-cracks increases with the increase of joint thickness. The change of joint angle seriously affects the development of micro-cracks and the final failure mode of rock， i.e.， when the joint angle is large， particularly with multiple joints， the overall stability of samples is poor， and samples are vulnerable to suffering sliding failure and cause serious damages.  
        关键词：filled rock joint;joint characteristics;particle flow simulation;mechanical properties;crack evolution   
        10
        |
        2
        |
        0
        <HTML>
        <L-PDF><Meta-XML>
        <引用本文> <批量引用> 38800153 false
        发布时间：2023-11-07
      • LIN Zequn,WU Haiou,YANG Zhenhua,ZHANG Zhi,WANG Dagang
        Vol. 62, Issue 5, Pages: 157-170(2023) DOI: 10.13471/j.cnki.acta.snus.2023D002
        摘要：The spatiotemporal distribution and variation of tropical cyclone precipitation have significant impacts on disaster-affected areas. However， studies on short duration tropical cyclone precipitation and its spatiotemporal distribution characteristics are lacking. Therefore， this study uses a precipitation product data with high spatial and temporal resolution， based on Gini index， Space Gini index and linear regression method， to study the spatial and temporal distribution characteristics and trends of tropical cyclone precipitation. The results show that： 1） the average annual impact frequency and annual precipitation of tropical cyclones decrease from southeast to northwest. The trend of influence frequency decreases significantly in most southeast regions of China， and the trend of annual precipitation shows obvious spatial differences. 2） Precipitation duration of tropical cyclones shows a decreasing trend， resulting in an increase in precipitation intensity and concentration， which indicates that tropical cyclone precipitation was more concentrated in time while its intensity is enhanced. 3） The analysis results of four tropical cyclones show that， within 200 km of the central path is the main precipitation area， which is uniform in temporal distribution. In the range of 800 km， precipitation is more concentrated in time with the increasing distance.  
        关键词：tropical cyclone precipitation;precipitation concentration;Gini index;spatiotemporal characteristics   
        23
        |
        1
        |
        0
        <HTML>
        <L-PDF><Meta-XML>
        <引用本文> <批量引用> 38800187 false
        发布时间：2023-11-07
      • CUI Yinping,JIN Yinbao,ZHANG Juan,SHEN Ao,WU Meng,TAN Haobo,LIU Yiming,LU Xiao,FAN Qi
        Vol. 62, Issue 5, Pages: 171-180(2023) DOI: 10.13471/j.cnki.acta.snus.2022D055
        摘要：The WRF mesoscale meteorological model and the HYSPLIT model were used to investigate a local radiation fog that occurred in Foshan from Nov 10th to Nov 11th， 2017， and its sources of water vapor. The results show this is a typical radiation fog process. During the clear night， surface long-wave radiation cooling resulted in water vapor gathering and condensing， meanwhile， temperature inversion and wind speed that lower than 1 m/s in the area promoted the development of the fog. Vertical temperature variations were highly related to water vapor concentrations， which could represent the features of vertical development during the fogging process. To analyze the sources of water vapor in this fog， high-resolution simulation results from the WRF model were utilized to drive the HYSPLIT model. It is found that in addition to the local water vapor condensation， the transport of water vapor from the Pearl River in Foshan city also promoted fogging.  
        关键词：WRF model;long wave radiation cooling;water vapor;HYSPLIT model   
        17
        |
        1
        |
        0
        <HTML>
        <L-PDF><Meta-XML>
        <引用本文> <批量引用> 38800225 false
        发布时间：2023-11-07
      • DONG Yanqiu,ZHAO Xiong,BAI Xuling,WANG Kaimin,DONG Xueyan,FAN Ruifeng,SUN Weiqing,MA Yulu
        Vol. 62, Issue 5, Pages: 181-186(2023) DOI: 10.13471/j.cnki.acta.snus.2023C001
        摘要：Under solvothermal conditions， a coordination polymer with fluorescence properties was constructed by using tricarboxylic acid derivative 5-（（4-carboxybenzyl） amino） isophthalic acid （H3L） and 4，4'-dimethyl-2，2'-bipyridine （dbpy） as mixed ligands with Cd（Ⅱ）， namely［Cd2（L）（dbpy）2（HCOO）（H2O）］•2H2O （CP 1）. Characterizing by X-ray single crystal diffraction， X-ray powder diffraction and thermogravimetric analysis， CP 1 was identified as a binuclear two-dimensional layered structure. In addition， CP 1 has good fluorescence， and can realize selective fluorescence quenching detection for hexavalent chromium oxylate ions （Cr2O72- and CrO42-） in aqueous solution.  
        关键词：coordination polymer;fluorescent sensor;chromium ion;dichromate detection;chromate detection   
        31
        |
        2
        |
        0
        <HTML>
        <L-PDF><Meta-XML>
        <引用本文> <批量引用> 38807570 false
        发布时间：2023-11-07
      0