A difference scheme in curvilinear coordinates is put forward for calculation of salinity in estuaries and coastal waters, which is based on Eulerian-Lagrangian method. It combines first-order and second-order Lagrang...A difference scheme in curvilinear coordinates is put forward for calculation of salinity in estuaries and coastal waters, which is based on Eulerian-Lagrangian method. It combines first-order and second-order Lagrangian interpolation to reduce numerical dispersion and oscillation. And the length of the curvilinear grid is also considered in the interpolation. Then the scheme is used in estuary, coast and ocean model, and several numerical experiments for the Yangtze Estuary and the Hangzhou Bay are conducted to test it. These experiments show that it is suitable for simulations of salinity in estuaries and coastal waters with the models using curvilinear coordinates.展开更多
In this paper,an efficien formulation based on the Lagrangian method is presented to investigate the contact–impact problems of f exible multi-body systems.Generally,the penalty method and the Hertz contact law are t...In this paper,an efficien formulation based on the Lagrangian method is presented to investigate the contact–impact problems of f exible multi-body systems.Generally,the penalty method and the Hertz contact law are the most commonly used methods in engineering applications.However,these methods are highly dependent on various non-physical parameters,which have great effects on the simulation results.Moreover,a tremendous number of degrees of freedom in the contact–impact problems will influenc thenumericalefficien ysignificantl.Withtheconsideration of these two problems,a formulation combining the component mode synthesis method and the Lagrangian method is presented to investigate the contact–impact problems in fl xible multi-body system numerically.Meanwhile,the finit element meshing laws of the contact bodies will be studied preliminarily.A numerical example with experimental verificatio will certify the reliability of the presented formulationincontact–impactanalysis.Furthermore,aseries of numerical investigations explain how great the influenc of the finit element meshing has on the simulation results.Finally the limitations of the element size in different regions are summarized to satisfy both the accuracy and efficien y.展开更多
The Eulerian？Lagrangian method（ELM） has been used by many ocean models as the solution of the advection equation,but the numerical error caused by interpolation imposes restriction on its accuracy.In the present st...The Eulerian？Lagrangian method（ELM） has been used by many ocean models as the solution of the advection equation,but the numerical error caused by interpolation imposes restriction on its accuracy.In the present study,hybrid N-order Lagrangian interpolation ELM（Li ELM） is put forward in which the N-order Lagrangian interpolation is used at first,then the lower order Lagrangian interpolation is applied in the points where the interpolation results are abnormally higher or lower.The calculation results of a step-shaped salinity advection model are analyzed,which show that higher order（N=3？8） Li ELM can reduce the mean numerical error of salinity calculation,but the numerical oscillation error is still significant.Even number order Li ELM makes larger numerical oscillation error than its adjacent odd number order Li ELM.Hybrid N-order Li ELM can remove numerical oscillation,and it significantly reduces the mean numerical error when N is even and the current is in fixed direction,while it makes less effect on mean numerical error when N is odd or the current direction changes periodically.Hybrid odd number order Li ELM makes less mean numerical error than its adjacent even number order Li ELM when the current is in the fixed direction,while the mean numerical error decreases as N increases when the current direction changes periodically,so odd number of N may be better for application.Among various types of Hybrid N-order Li ELM,the scheme reducing N-order directly to 1st-order may be the optimal for synthetic selection of accuracy and computational efficiency.展开更多
This paper formulates a two-dimensional strip packing problem as a non- linear programming （NLP） problem and establishes the first-order optimality conditions for the NLP problem. A numerical algorithm for solving t...This paper formulates a two-dimensional strip packing problem as a non- linear programming （NLP） problem and establishes the first-order optimality conditions for the NLP problem. A numerical algorithm for solving this NLP problem is given to find exact solutions to strip-packing problems involving up to 10 items. Approximate solutions can be found for big-sized problems by decomposing the set of items into small-sized blocks of which each block adopts the proposed numerical algorithm. Numerical results show that the approximate solutions to big-sized problems obtained by this method are superior to those by NFDH, FFDH and BFDH approaches.展开更多
A large number of problems in engineering can be formulated as the optimization of certain functionals. In this paper, we present an algorithm that uses the augmented Lagrangian methods for finding numerical solutions...A large number of problems in engineering can be formulated as the optimization of certain functionals. In this paper, we present an algorithm that uses the augmented Lagrangian methods for finding numerical solutions to engineering problems. These engineering problems are described by differential equations with boundary values and are formulated as optimization of some functionals. The algorithm achieves its simplicity and versatility by choosing linear equality relations recursively for the augmented Lagrangian associated with an optimization problem. We demonstrate the formulation of an optimization functional for a 4th order nonlinear differential equation with boundary values. We also derive the associated augmented Lagrangian for this 4th order differential equation. Numerical test results are included that match up with well-established experimental outcomes. These numerical results indicate that the new algorithm is fully capable of producing accurate and stable solutions to differential equations.展开更多
In this paper,we analyze the convergence properties of a stochastic augmented Lagrangian method for solving stochastic convex programming problems with inequality constraints.Approximation models for stochastic convex...In this paper,we analyze the convergence properties of a stochastic augmented Lagrangian method for solving stochastic convex programming problems with inequality constraints.Approximation models for stochastic convex programming problems are constructed from stochastic observations of real objective and constraint functions.Based on relations between solutions of the primal problem and solutions of the dual problem,it is proved that the convergence of the algorithm from the perspective of the dual problem.Without assumptions on how these random models are generated,when estimates are merely sufficiently accurate to the real objective and constraint functions with high enough,but fixed,probability,the method converges globally to the optimal solution almost surely.In addition,sufficiently accurate random models are given under different noise assumptions.We also report numerical results that show the good performance of the algorithm for different convex programming problems with several random models.展开更多
The augmented Lagrangian function and the corresponding augmented Lagrangian method are constructed for solving a class of minimax optimization problems with equality constraints.We prove that,under the linear indepen...The augmented Lagrangian function and the corresponding augmented Lagrangian method are constructed for solving a class of minimax optimization problems with equality constraints.We prove that,under the linear independence constraint qualification and the second-order sufficiency optimality condition for the lower level problem and the second-order sufficiency optimality condition for the minimax problem,for a given multiplier vectorμ,the rate of convergence of the augmented Lagrangian method is linear with respect to||μu-μ^(*)||and the ratio constant is proportional to 1/c when the ratio|μ-μ^(*)||/c is small enough,where c is the penalty parameter that exceeds a threshold c_(*)>O andμ^(*)is the multiplier corresponding to a local minimizer.Moreover,we prove that the sequence of multiplier vectors generated by the augmented Lagrangian method has at least Q-linear convergence if the sequence of penalty parameters(ck)is bounded and the convergence rate is superlinear if(ck)is increasing to infinity.Finally,we use a direct way to establish the rate of convergence of the augmented Lagrangian method for the minimax problem with a quadratic objective function and linear equality constraints.展开更多
The seasonal variations of the Kuroshio intrusion pathways northeast of Taiwan were investigated using observational data from satellite-tracked sea surface drifters and a numerical particle-tracking experiment based ...The seasonal variations of the Kuroshio intrusion pathways northeast of Taiwan were investigated using observational data from satellite-tracked sea surface drifters and a numerical particle-tracking experiment based on a high-resolution numerical ocean model. The results of sea surface drifter data observed from 1989 to 2013 indicate that the Kuroshio surface intrusion follows two distinct pathways: one is a northwestward intrusion along the northern coast of Taiwan Island, and the other is a direct intrusion near the turn of the shelf break. The former occurs primarily in the winter, while the latter exists year round. A particle-tracking experiment in the high-resolution numerical model reproduces the two observed intrusion paths by the sea surface drifters. The three-dimensional structure of the Kuroshio intrusion is revealed by the model results. The pathways, features and possible dynamic mechanisms of the subsurface intrusion are also discussed.展开更多
Transports of air particulate matters(PM) from face sources in the atmospheric boundary layer(ABL) are investigated by the Eulerian single fluid model and the Lagrangian trajectory method,respectively.Large eddy simul...Transports of air particulate matters(PM) from face sources in the atmospheric boundary layer(ABL) are investigated by the Eulerian single fluid model and the Lagrangian trajectory method,respectively.Large eddy simulation is used to simulate the fluid phase for high accuracy in both two approaches.The mean and fluctuating PM concentrations,as well as instantaneous PM distributions at different downstream and height positions,are presented.Higher mean and fluctuating particle concentrations are predicted by the Eulerian approach than the Lagrangian one.For the Lagrangian method,PM distributions cluster near the ground-wall because of the preferential dispersion of inertial particles by turbulence structures in the ABL,while it cannot be obtained by the Eulerian single fluid method,because the two-phase velocity differences are neglected in the Eulerian method.展开更多
This paper provides an analysis on the effects of exact and inexact integrations on stability, convergence, numerical diffusion, and numerical oscillations for the Eulerian- Lagrangian method （ELM）. In the finite el...This paper provides an analysis on the effects of exact and inexact integrations on stability, convergence, numerical diffusion, and numerical oscillations for the Eulerian- Lagrangian method （ELM）. In the finite element ELM, when more accurate integrations are used for the right-hand-side, less numerical diffusion is introduced and better approximation is obtained. When linear interpolation is used for numerical integrations, the resulting ELM is shown to be unconditionally stable and of first-order accuracy. When Gauss quadrature is used, conditional stability and second-order accuracy are established under some mild constraints for the convection-diffusion problems. Finally, numerical experiments demonstrate that more accurate integrations lead to better approximation, and spatial adaptivity can substantially reduce numerical oscillations and smearing that often occur in the ELM when inexact numerical integrations are used.展开更多
Based on the conservation of entropy and potential vorticity in adiabatic atmospheric motion without the consideration of friction,calculation is made of the trajectory of a particle on an isentropic surface by use of...Based on the conservation of entropy and potential vorticity in adiabatic atmospheric motion without the consideration of friction,calculation is made of the trajectory of a particle on an isentropic surface by use of the data of FGGE III-b.Results of several calculation schemes of the trajectory discussed show that the local data interpolation and Runge-Kutta time-integral scheme is the best.The calculated trajectory reflects the large-scale atmospheric motion only and the small-scale motion emerges as a deviation term of the calculated trajectory.And then the outbreak and propagation of planetary wave are studied by means of the deformation of a material line,with the result showing that the material line can be tracked in the trop- osphere only in a few days,beyond which the interaction between the small-scale waves and large-scale motion leads to its dramatical twisting and deformation.Therefore,the Lagrangian method is assumed to be an effective means of diagnostic research in the nonlinear intcraction in atmospheric circulation,in addition to the general study of the atmospheric circulation.展开更多
In this paper,we provide some gentle introductions to the recent advance in augmented Lagrangian methods for solving large-scale convex matrix optimization problems(cMOP).Specifically,we reviewed two types of sufficie...In this paper,we provide some gentle introductions to the recent advance in augmented Lagrangian methods for solving large-scale convex matrix optimization problems(cMOP).Specifically,we reviewed two types of sufficient conditions for ensuring the quadratic growth conditions of a class of constrained convex matrix optimization problems regularized by nonsmooth spectral functions.Under a mild quadratic growth condition on the dual of cMOP,we further discussed the R-superlinear convergence of the Karush-Kuhn-Tucker(KKT)residuals of the sequence generated by the augmented Lagrangian methods(ALM)for solving convex matrix optimization problems.Implementation details of the ALM for solving core convex matrix optimization problems are also provided.展开更多
In this paper, a Petrov-Galerkin scheme named the Runge-Kutta control volume （RKCV） discontinuous finite ele- ment method is constructed to solve the one-dimensional compressible Euler equations in the Lagrangian co...In this paper, a Petrov-Galerkin scheme named the Runge-Kutta control volume （RKCV） discontinuous finite ele- ment method is constructed to solve the one-dimensional compressible Euler equations in the Lagrangian coordinate. Its advantages include preservation of the local conservation and a high resolution. Compared with the Runge-Kutta discon- tinuous Galerkin （RKDG） method, the RKCV method is easier to implement. Moreover, the advantages of the RKCV and the Lagrangian methods are combined in the new method. Several numerical examples are given to illustrate the accuracy and the reliability of the algorithm.展开更多
An Arbitrary Lagrangian-Eulerian(ALE) method was employed to simulate the sheet metal extrusion process,aiming at avoiding mesh distortion and improving the computational accuracy.The method was implemented based on M...An Arbitrary Lagrangian-Eulerian(ALE) method was employed to simulate the sheet metal extrusion process,aiming at avoiding mesh distortion and improving the computational accuracy.The method was implemented based on MSC/MARC by using a fractional step method,i.e.a Lagrangian step followed by an Euler step.The Lagrangian step was a pure updated Lagrangian calculation and the Euler step was performed using mesh smoothing and remapping scheme.Due to the extreme distortion of deformation domain,it was almost impossible to complete the whole simulation with only one mesh topology.Therefore,global remeshing combined with the ALE method was used in the simulation work.Based on the numerical model of the process,some deformation features of the sheet metal extrusion process,such as distribution of localized equivalent plastic strain,and shrinkage cavity,were revealed.Furthermore,the differences between conventional extrusion and sheet metal extrusion process were also analyzed.展开更多
Tensor robust principal component analysis（TRPCA） problem aims to separate a low-rank tensor and a sparse tensor from their sum. This problem has recently attracted considerable research attention due to its wide ra...Tensor robust principal component analysis（TRPCA） problem aims to separate a low-rank tensor and a sparse tensor from their sum. This problem has recently attracted considerable research attention due to its wide range of potential applications in computer vision and pattern recognition. In this paper, we propose a new model to deal with the TRPCA problem by an alternation minimization algorithm along with two adaptive rankadjusting strategies. For the underlying low-rank tensor, we simultaneously perform low-rank matrix factorizations to its all-mode matricizations; while for the underlying sparse tensor,a soft-threshold shrinkage scheme is applied. Our method can be used to deal with the separation between either an exact or an approximate low-rank tensor and a sparse one. We established the subsequence convergence of our algorithm in the sense that any limit point of the iterates satisfies the KKT conditions. When the iteration stops, the output will be modified by applying a high-order SVD approach to achieve an exactly low-rank final result as the accurate rank has been calculated. The numerical experiments demonstrate that our method could achieve better results than the compared methods.展开更多
In this paper,Runge-Kutta Discontinuous Galerkin（RKDG） finite element method is presented to solve the onedimensional inviscid compressible gas dynamic equations in a Lagrangian coordinate.The equations are discreti...In this paper,Runge-Kutta Discontinuous Galerkin（RKDG） finite element method is presented to solve the onedimensional inviscid compressible gas dynamic equations in a Lagrangian coordinate.The equations are discretized by the DG method in space and the temporal discretization is accomplished by the total variation diminishing Runge-Kutta method.A limiter based on the characteristic field decomposition is applied to maintain stability and non-oscillatory property of the RKDG method.For multi-medium fluid simulation,the two cells adjacent to the interface are treated differently from other cells.At first,a linear Riemann solver is applied to calculate the numerical ？ux at the interface.Numerical examples show that there is some oscillation in the vicinity of the interface.Then a nonlinear Riemann solver based on the characteristic formulation of the equation and the discontinuity relations is adopted to calculate the numerical ？ux at the interface,which suppresses the oscillation successfully.Several single-medium and multi-medium fluid examples are given to demonstrate the reliability and efficiency of the algorithm.展开更多
The numerical approximation of the Spectral-Lagrangian scheme developed by the authors in [30] for a wide range of homogeneous non-linear Boltzmann type equations is extended to the space inhomogeneous case and severa...The numerical approximation of the Spectral-Lagrangian scheme developed by the authors in [30] for a wide range of homogeneous non-linear Boltzmann type equations is extended to the space inhomogeneous case and several shock problems are benchmark. Recognizing that the Boltzmann equation is an important tool in the analysis of formation of shock and boundary layer structures, we present the computational algorithm in Section 3.3 and perform a numerical study case in shock tube geometries well modeled in for ID in x times 3D in v in Section 4. The classic Riemann problem is numerically analyzed for Knudsen numbers close to continuum. The shock tube problem of Aoki et al [2], where the wall temperature is suddenly increased or decreased, is also studied. We consider the problem of heat transfer between two parallel plates with diffusive boundary conditions for a range of Knudsen numbers from close to continuum to a highly rarefied state. Finally, the classical infinite shock tube problem that generates a non-moving shock wave is studied. The point worth noting in this example is that the flow in the final case turns from a supersonic flow to a subsonic flow across the shock.展开更多
In the Lagrangian meshless(particle)methods,such as the smoothed particle hydrodynamics(SPH),moving particle semi-implicit(MPS)method and meshless local Petrov-Galerkin method based on Rankine source solution(MLPG_R),...In the Lagrangian meshless(particle)methods,such as the smoothed particle hydrodynamics(SPH),moving particle semi-implicit(MPS)method and meshless local Petrov-Galerkin method based on Rankine source solution(MLPG_R),the Laplacian discretisation is often required in order to solve the governing equations and/or estimate physical quantities(such as the viscous stresses).In some meshless applications,the Laplacians are also needed as stabilisation operators to enhance the pressure calculation.The particles in the Lagrangian methods move following the material velocity,yielding a disordered(random)particle distribution even though they may be distributed uniformly in the initial state.Different schemes have been developed for a direct estimation of second derivatives using finite difference,kernel integrations and weighted/moving least square method.Some of the schemes suffer from a poor convergent rate.Some have a better convergent rate but require inversions of high order matrices,yielding high computational costs.This paper presents a quadric semi-analytical finite-difference interpolation(QSFDI)scheme,which can achieve the same degree of the convergent rate as the best schemes available to date but requires the inversion of significant lower-order matrices,i.e.3×3 for 3D cases,compared with 6×6 or 10×10 in the schemes with the best convergent rate.Systematic patch tests have been carried out for either estimating the Laplacian of given functions or solving Poisson’s equations.The convergence,accuracy and robustness of the present schemes are compared with the existing schemes.It will show that the present scheme requires considerably less computational time to achieve the same accuracy as the best schemes available in literatures,particularly for estimating the Laplacian of given functions.展开更多
The classical Lagrangian particle tracing method is widely used in the evaluation of the ocean annual subduction rate.However,our analysis indicates that in addition to neglecting the effect of mixing,there are two po...The classical Lagrangian particle tracing method is widely used in the evaluation of the ocean annual subduction rate.However,our analysis indicates that in addition to neglecting the effect of mixing,there are two possible deviations in the method:one is an overestimation due to not considering that the amount of subducted water at the source location may be inadequate during the late winter of the first year when the mixed layer becomes shallow;the other one is an underestimation due to the neglect of the effective subduction caused by strong vertical pumping.Quantitative analysis shows that these two deviations mainly exist in the low-latitude subduction areas of the South Pacific and South Atlantic.The two deviations have very similar distribution areas and can partially off set each other.However,the overall deviation is still large,and the maximum relative deviation ratio can reach 50%;therefore,it cannot be ignored.展开更多
A new approach for treating the mesh with Lagrangian scheme of finite volume method is presented. It has been proved that classical Lagrangian method is difficult to cope with large deformation in tracking material pa...A new approach for treating the mesh with Lagrangian scheme of finite volume method is presented. It has been proved that classical Lagrangian method is difficult to cope with large deformation in tracking material particles due to severe distortion of cells, and the changing connectivity of the mesh seems especially attractive for solving such issues. The mesh with large deformation based on computational geometry is optimized by using new method. This paper develops a processing system for arbitrary polygonal unstructured grid,the intelligent variable grid neighborhood technologies is utilized to improve the quality of mesh in calculation process, and arbitrary polygonal mesh is used in the Lagrangian finite volume scheme. The performance of the new method is demonstrated through series of numerical examples, and the simulation capability is efficiently presented in coping with the systems with large deformations.展开更多
基金This project was supported by the Major State Basic Research Program under Contract Grant No. G1999043803the University Fund for Mainstay Teachers of State Ministry of Education and the Opening Fund of Open Laboratory of Marine Dynamic Process and Sa
文摘A difference scheme in curvilinear coordinates is put forward for calculation of salinity in estuaries and coastal waters, which is based on Eulerian-Lagrangian method. It combines first-order and second-order Lagrangian interpolation to reduce numerical dispersion and oscillation. And the length of the curvilinear grid is also considered in the interpolation. Then the scheme is used in estuary, coast and ocean model, and several numerical experiments for the Yangtze Estuary and the Hangzhou Bay are conducted to test it. These experiments show that it is suitable for simulations of salinity in estuaries and coastal waters with the models using curvilinear coordinates.
基金supported by the National Science Foundation of China (Grants 11132007,11272203)
文摘In this paper,an efficien formulation based on the Lagrangian method is presented to investigate the contact–impact problems of f exible multi-body systems.Generally,the penalty method and the Hertz contact law are the most commonly used methods in engineering applications.However,these methods are highly dependent on various non-physical parameters,which have great effects on the simulation results.Moreover,a tremendous number of degrees of freedom in the contact–impact problems will influenc thenumericalefficien ysignificantl.Withtheconsideration of these two problems,a formulation combining the component mode synthesis method and the Lagrangian method is presented to investigate the contact–impact problems in fl xible multi-body system numerically.Meanwhile,the finit element meshing laws of the contact bodies will be studied preliminarily.A numerical example with experimental verificatio will certify the reliability of the presented formulationincontact–impactanalysis.Furthermore,aseries of numerical investigations explain how great the influenc of the finit element meshing has on the simulation results.Finally the limitations of the element size in different regions are summarized to satisfy both the accuracy and efficien y.
基金financially supported by the National Natural Science Foundation of China(Grant Nos.40906044 and 41076048)the Fundamental Research Funds for the Central Universities Project(Grant No.2011B05714)
文摘The Eulerian？Lagrangian method（ELM） has been used by many ocean models as the solution of the advection equation,but the numerical error caused by interpolation imposes restriction on its accuracy.In the present study,hybrid N-order Lagrangian interpolation ELM（Li ELM） is put forward in which the N-order Lagrangian interpolation is used at first,then the lower order Lagrangian interpolation is applied in the points where the interpolation results are abnormally higher or lower.The calculation results of a step-shaped salinity advection model are analyzed,which show that higher order（N=3？8） Li ELM can reduce the mean numerical error of salinity calculation,but the numerical oscillation error is still significant.Even number order Li ELM makes larger numerical oscillation error than its adjacent odd number order Li ELM.Hybrid N-order Li ELM can remove numerical oscillation,and it significantly reduces the mean numerical error when N is even and the current is in fixed direction,while it makes less effect on mean numerical error when N is odd or the current direction changes periodically.Hybrid odd number order Li ELM makes less mean numerical error than its adjacent even number order Li ELM when the current is in the fixed direction,while the mean numerical error decreases as N increases when the current direction changes periodically,so odd number of N may be better for application.Among various types of Hybrid N-order Li ELM,the scheme reducing N-order directly to 1st-order may be the optimal for synthetic selection of accuracy and computational efficiency.
基金State Foundstion of Ph.D Units of China(2003-05)under Grant 20020141013the NNSF(10471015)of Liaoning Province,China.
文摘This paper formulates a two-dimensional strip packing problem as a non- linear programming （NLP） problem and establishes the first-order optimality conditions for the NLP problem. A numerical algorithm for solving this NLP problem is given to find exact solutions to strip-packing problems involving up to 10 items. Approximate solutions can be found for big-sized problems by decomposing the set of items into small-sized blocks of which each block adopts the proposed numerical algorithm. Numerical results show that the approximate solutions to big-sized problems obtained by this method are superior to those by NFDH, FFDH and BFDH approaches.
文摘A large number of problems in engineering can be formulated as the optimization of certain functionals. In this paper, we present an algorithm that uses the augmented Lagrangian methods for finding numerical solutions to engineering problems. These engineering problems are described by differential equations with boundary values and are formulated as optimization of some functionals. The algorithm achieves its simplicity and versatility by choosing linear equality relations recursively for the augmented Lagrangian associated with an optimization problem. We demonstrate the formulation of an optimization functional for a 4th order nonlinear differential equation with boundary values. We also derive the associated augmented Lagrangian for this 4th order differential equation. Numerical test results are included that match up with well-established experimental outcomes. These numerical results indicate that the new algorithm is fully capable of producing accurate and stable solutions to differential equations.
基金supported by the National Key R&D Program of China(Project No.2022YFA1004000)by the Natural Science Foundation of China(Project No.12371298).
文摘In this paper,we analyze the convergence properties of a stochastic augmented Lagrangian method for solving stochastic convex programming problems with inequality constraints.Approximation models for stochastic convex programming problems are constructed from stochastic observations of real objective and constraint functions.Based on relations between solutions of the primal problem and solutions of the dual problem,it is proved that the convergence of the algorithm from the perspective of the dual problem.Without assumptions on how these random models are generated,when estimates are merely sufficiently accurate to the real objective and constraint functions with high enough,but fixed,probability,the method converges globally to the optimal solution almost surely.In addition,sufficiently accurate random models are given under different noise assumptions.We also report numerical results that show the good performance of the algorithm for different convex programming problems with several random models.
基金the National Natural Science Foundation of China(Nos.11991020,11631013,11971372,11991021,11971089 and 11731013)the Strategic Priority Research Program of Chinese Academy of Sciences(No.XDA27000000)Dalian High-Level Talent Innovation Project(No.2020RD09)。
文摘The augmented Lagrangian function and the corresponding augmented Lagrangian method are constructed for solving a class of minimax optimization problems with equality constraints.We prove that,under the linear independence constraint qualification and the second-order sufficiency optimality condition for the lower level problem and the second-order sufficiency optimality condition for the minimax problem,for a given multiplier vectorμ,the rate of convergence of the augmented Lagrangian method is linear with respect to||μu-μ^(*)||and the ratio constant is proportional to 1/c when the ratio|μ-μ^(*)||/c is small enough,where c is the penalty parameter that exceeds a threshold c_(*)>O andμ^(*)is the multiplier corresponding to a local minimizer.Moreover,we prove that the sequence of multiplier vectors generated by the augmented Lagrangian method has at least Q-linear convergence if the sequence of penalty parameters(ck)is bounded and the convergence rate is superlinear if(ck)is increasing to infinity.Finally,we use a direct way to establish the rate of convergence of the augmented Lagrangian method for the minimax problem with a quadratic objective function and linear equality constraints.
基金supported by the National Basic Research Program of China(Grant Nos.2013CB430302&2014CB-745000)the National Natural Science Foundation of China(Grant Nos.41321004+5 种基金911282044110602041476022&41490643)the Scientific Research Fund of the Second Institute of OceanographySOA(Grant No.JT1205)the National Program on Global Change and Air-Sea Interaction(Grant No.GASI-03-IPOVAI-05)
文摘The seasonal variations of the Kuroshio intrusion pathways northeast of Taiwan were investigated using observational data from satellite-tracked sea surface drifters and a numerical particle-tracking experiment based on a high-resolution numerical ocean model. The results of sea surface drifter data observed from 1989 to 2013 indicate that the Kuroshio surface intrusion follows two distinct pathways: one is a northwestward intrusion along the northern coast of Taiwan Island, and the other is a direct intrusion near the turn of the shelf break. The former occurs primarily in the winter, while the latter exists year round. A particle-tracking experiment in the high-resolution numerical model reproduces the two observed intrusion paths by the sea surface drifters. The three-dimensional structure of the Kuroshio intrusion is revealed by the model results. The pathways, features and possible dynamic mechanisms of the subsurface intrusion are also discussed.
基金supported by the National Natural Science Foundation of China (Grant Nos. 50876053 and 11132005)Opening fund of State of Key Laboratory of Nonlinear Mechanics
文摘Transports of air particulate matters(PM) from face sources in the atmospheric boundary layer(ABL) are investigated by the Eulerian single fluid model and the Lagrangian trajectory method,respectively.Large eddy simulation is used to simulate the fluid phase for high accuracy in both two approaches.The mean and fluctuating PM concentrations,as well as instantaneous PM distributions at different downstream and height positions,are presented.Higher mean and fluctuating particle concentrations are predicted by the Eulerian approach than the Lagrangian one.For the Lagrangian method,PM distributions cluster near the ground-wall because of the preferential dispersion of inertial particles by turbulence structures in the ABL,while it cannot be obtained by the Eulerian single fluid method,because the two-phase velocity differences are neglected in the Eulerian method.
文摘This paper provides an analysis on the effects of exact and inexact integrations on stability, convergence, numerical diffusion, and numerical oscillations for the Eulerian- Lagrangian method （ELM）. In the finite element ELM, when more accurate integrations are used for the right-hand-side, less numerical diffusion is introduced and better approximation is obtained. When linear interpolation is used for numerical integrations, the resulting ELM is shown to be unconditionally stable and of first-order accuracy. When Gauss quadrature is used, conditional stability and second-order accuracy are established under some mild constraints for the convection-diffusion problems. Finally, numerical experiments demonstrate that more accurate integrations lead to better approximation, and spatial adaptivity can substantially reduce numerical oscillations and smearing that often occur in the ELM when inexact numerical integrations are used.
文摘Based on the conservation of entropy and potential vorticity in adiabatic atmospheric motion without the consideration of friction,calculation is made of the trajectory of a particle on an isentropic surface by use of the data of FGGE III-b.Results of several calculation schemes of the trajectory discussed show that the local data interpolation and Runge-Kutta time-integral scheme is the best.The calculated trajectory reflects the large-scale atmospheric motion only and the small-scale motion emerges as a deviation term of the calculated trajectory.And then the outbreak and propagation of planetary wave are studied by means of the deformation of a material line,with the result showing that the material line can be tracked in the trop- osphere only in a few days,beyond which the interaction between the small-scale waves and large-scale motion leads to its dramatical twisting and deformation.Therefore,the Lagrangian method is assumed to be an effective means of diagnostic research in the nonlinear intcraction in atmospheric circulation,in addition to the general study of the atmospheric circulation.
基金Chao Ding’s research was supported by the National Natural Science Foundation of China(Nos.11671387,11531014,and 11688101)Beijing Natural Science Foundation(No.Z190002)+6 种基金Xu-Dong Li’s research was supported by the National Key R&D Program of China(No.2020YFA0711900)the National Natural Science Foundation of China(No.11901107)the Young Elite Scientists Sponsorship Program by CAST(No.2019QNRC001)the Shanghai Sailing Program(No.19YF1402600)the Science and Technology Commission of Shanghai Municipality Project(No.19511120700)Xin-Yuan Zhao’s research was supported by the National Natural Science Foundation of China(No.11871002)the General Program of Science and Technology of Beijing Municipal Education Commission(No.KM201810005004).
文摘In this paper,we provide some gentle introductions to the recent advance in augmented Lagrangian methods for solving large-scale convex matrix optimization problems(cMOP).Specifically,we reviewed two types of sufficient conditions for ensuring the quadratic growth conditions of a class of constrained convex matrix optimization problems regularized by nonsmooth spectral functions.Under a mild quadratic growth condition on the dual of cMOP,we further discussed the R-superlinear convergence of the Karush-Kuhn-Tucker(KKT)residuals of the sequence generated by the augmented Lagrangian methods(ALM)for solving convex matrix optimization problems.Implementation details of the ALM for solving core convex matrix optimization problems are also provided.
基金Project supported by the National Natural Science Foundation of China (Grant Nos. 11261035 and 11171038)the Science Research Foundation of the Institute of Higher Education of Inner Mongolia Autonomous Region, China (Grant No. NJZZ12198)the Natural Science Foundation of Inner Mongolia Autonomous Region, China (Grant No. 2012MS0102)
文摘In this paper, a Petrov-Galerkin scheme named the Runge-Kutta control volume （RKCV） discontinuous finite ele- ment method is constructed to solve the one-dimensional compressible Euler equations in the Lagrangian coordinate. Its advantages include preservation of the local conservation and a high resolution. Compared with the Runge-Kutta discon- tinuous Galerkin （RKDG） method, the RKCV method is easier to implement. Moreover, the advantages of the RKCV and the Lagrangian methods are combined in the new method. Several numerical examples are given to illustrate the accuracy and the reliability of the algorithm.
基金Project(50505027) supported by the National Natural Science Foundation of ChinaProject(20070248056) supported by the Research Fund for the Doctoral Program of Higher Education of China
文摘An Arbitrary Lagrangian-Eulerian(ALE) method was employed to simulate the sheet metal extrusion process,aiming at avoiding mesh distortion and improving the computational accuracy.The method was implemented based on MSC/MARC by using a fractional step method,i.e.a Lagrangian step followed by an Euler step.The Lagrangian step was a pure updated Lagrangian calculation and the Euler step was performed using mesh smoothing and remapping scheme.Due to the extreme distortion of deformation domain,it was almost impossible to complete the whole simulation with only one mesh topology.Therefore,global remeshing combined with the ALE method was used in the simulation work.Based on the numerical model of the process,some deformation features of the sheet metal extrusion process,such as distribution of localized equivalent plastic strain,and shrinkage cavity,were revealed.Furthermore,the differences between conventional extrusion and sheet metal extrusion process were also analyzed.
基金Supported by the National Natural Science Foundation of China(Grant Nos.6157209961320106008+2 种基金91230103)National Science and Technology Major Project(Grant Nos.2013ZX040050212014ZX04001011)
文摘Tensor robust principal component analysis（TRPCA） problem aims to separate a low-rank tensor and a sparse tensor from their sum. This problem has recently attracted considerable research attention due to its wide range of potential applications in computer vision and pattern recognition. In this paper, we propose a new model to deal with the TRPCA problem by an alternation minimization algorithm along with two adaptive rankadjusting strategies. For the underlying low-rank tensor, we simultaneously perform low-rank matrix factorizations to its all-mode matricizations; while for the underlying sparse tensor,a soft-threshold shrinkage scheme is applied. Our method can be used to deal with the separation between either an exact or an approximate low-rank tensor and a sparse one. We established the subsequence convergence of our algorithm in the sense that any limit point of the iterates satisfies the KKT conditions. When the iteration stops, the output will be modified by applying a high-order SVD approach to achieve an exactly low-rank final result as the accurate rank has been calculated. The numerical experiments demonstrate that our method could achieve better results than the compared methods.
基金Project supported by the National Natural Science Foundation of China (Grant Nos. 11261035,11171038,and 10771019)the Science Reaearch Foundation of Institute of Higher Education of Inner Mongolia Autonomous Region,China (Grant No. NJZZ12198)the Natural Science Foundation of Inner Mongolia Autonomous Region,China (Grant No. 2012MS0102)
文摘In this paper,Runge-Kutta Discontinuous Galerkin（RKDG） finite element method is presented to solve the onedimensional inviscid compressible gas dynamic equations in a Lagrangian coordinate.The equations are discretized by the DG method in space and the temporal discretization is accomplished by the total variation diminishing Runge-Kutta method.A limiter based on the characteristic field decomposition is applied to maintain stability and non-oscillatory property of the RKDG method.For multi-medium fluid simulation,the two cells adjacent to the interface are treated differently from other cells.At first,a linear Riemann solver is applied to calculate the numerical ？ux at the interface.Numerical examples show that there is some oscillation in the vicinity of the interface.Then a nonlinear Riemann solver based on the characteristic formulation of the equation and the discontinuity relations is adopted to calculate the numerical ？ux at the interface,which suppresses the oscillation successfully.Several single-medium and multi-medium fluid examples are given to demonstrate the reliability and efficiency of the algorithm.
基金partially supported under the NSF grant DMS-0507038 and 0807712Support from the Institute of Computational Engineering and Sciences
文摘The numerical approximation of the Spectral-Lagrangian scheme developed by the authors in [30] for a wide range of homogeneous non-linear Boltzmann type equations is extended to the space inhomogeneous case and several shock problems are benchmark. Recognizing that the Boltzmann equation is an important tool in the analysis of formation of shock and boundary layer structures, we present the computational algorithm in Section 3.3 and perform a numerical study case in shock tube geometries well modeled in for ID in x times 3D in v in Section 4. The classic Riemann problem is numerically analyzed for Knudsen numbers close to continuum. The shock tube problem of Aoki et al [2], where the wall temperature is suddenly increased or decreased, is also studied. We consider the problem of heat transfer between two parallel plates with diffusive boundary conditions for a range of Knudsen numbers from close to continuum to a highly rarefied state. Finally, the classical infinite shock tube problem that generates a non-moving shock wave is studied. The point worth noting in this example is that the flow in the final case turns from a supersonic flow to a subsonic flow across the shock.
文摘In the Lagrangian meshless(particle)methods,such as the smoothed particle hydrodynamics(SPH),moving particle semi-implicit(MPS)method and meshless local Petrov-Galerkin method based on Rankine source solution(MLPG_R),the Laplacian discretisation is often required in order to solve the governing equations and/or estimate physical quantities(such as the viscous stresses).In some meshless applications,the Laplacians are also needed as stabilisation operators to enhance the pressure calculation.The particles in the Lagrangian methods move following the material velocity,yielding a disordered(random)particle distribution even though they may be distributed uniformly in the initial state.Different schemes have been developed for a direct estimation of second derivatives using finite difference,kernel integrations and weighted/moving least square method.Some of the schemes suffer from a poor convergent rate.Some have a better convergent rate but require inversions of high order matrices,yielding high computational costs.This paper presents a quadric semi-analytical finite-difference interpolation(QSFDI)scheme,which can achieve the same degree of the convergent rate as the best schemes available to date but requires the inversion of significant lower-order matrices,i.e.3×3 for 3D cases,compared with 6×6 or 10×10 in the schemes with the best convergent rate.Systematic patch tests have been carried out for either estimating the Laplacian of given functions or solving Poisson’s equations.The convergence,accuracy and robustness of the present schemes are compared with the existing schemes.It will show that the present scheme requires considerably less computational time to achieve the same accuracy as the best schemes available in literatures,particularly for estimating the Laplacian of given functions.
基金Supported by the National Natural Science Foundation of China(No.41676009)the National Key R&D Program of China(No.2016YFC0301203)the State Key Program of National Natural Science of China(No.41730534)。
文摘The classical Lagrangian particle tracing method is widely used in the evaluation of the ocean annual subduction rate.However,our analysis indicates that in addition to neglecting the effect of mixing,there are two possible deviations in the method:one is an overestimation due to not considering that the amount of subducted water at the source location may be inadequate during the late winter of the first year when the mixed layer becomes shallow;the other one is an underestimation due to the neglect of the effective subduction caused by strong vertical pumping.Quantitative analysis shows that these two deviations mainly exist in the low-latitude subduction areas of the South Pacific and South Atlantic.The two deviations have very similar distribution areas and can partially off set each other.However,the overall deviation is still large,and the maximum relative deviation ratio can reach 50%;therefore,it cannot be ignored.
基金supported in part by the National Natural Science Foundation of China under Grant 11372051,Grant 11475029part by the Fund of the China Academy of Engineering Physics under Grant 20150202045
文摘A new approach for treating the mesh with Lagrangian scheme of finite volume method is presented. It has been proved that classical Lagrangian method is difficult to cope with large deformation in tracking material particles due to severe distortion of cells, and the changing connectivity of the mesh seems especially attractive for solving such issues. The mesh with large deformation based on computational geometry is optimized by using new method. This paper develops a processing system for arbitrary polygonal unstructured grid,the intelligent variable grid neighborhood technologies is utilized to improve the quality of mesh in calculation process, and arbitrary polygonal mesh is used in the Lagrangian finite volume scheme. The performance of the new method is demonstrated through series of numerical examples, and the simulation capability is efficiently presented in coping with the systems with large deformations.
