The difficulty in solving stochastic dynamics problems lies in the need for a large number of repeated computations of deterministic dynamic equations,which has been a challenge in stochastic dynamics analysis and was...The difficulty in solving stochastic dynamics problems lies in the need for a large number of repeated computations of deterministic dynamic equations,which has been a challenge in stochastic dynamics analysis and was discussed in this study.To efficiently and accurately compute the exponential of the dynamics state matrix and the matrix functions due to external loads,an adaptively filtered precise integration method was proposed,which inherits the high precision of the precise integrationmethod,improves the computational efficiency and saves the memory required.Moreover,the perturbation method was introduced to avoid repeated computations of matrix exponential and terms due to external loads.Based on the filtering and perturbation techniques,an adaptively filtered precise integration method considering perturbation for stochastic dynamics problems was developed.Two numerical experiments,including a model of phononic crystal and a bridge model considering random parameters,were performed to test the performance of the proposed method in terms of accuracy and efficiency.Numerical results show that the accuracy and efficiency of the proposed method are better than those of the existing precise integration method,the Newmark-βmethod and the Wilson-θmethod.展开更多
The high-precision local geoid model was computed based on the improved Stokes-Helmert0 s boundary value problem and strict integrals of topographic effects. This proposed method involves three steps.First, the mathem...The high-precision local geoid model was computed based on the improved Stokes-Helmert0 s boundary value problem and strict integrals of topographic effects. This proposed method involves three steps.First, the mathematical form of Stokes-Helmert0 s boundary value problem was derived, and strict computational formulas regarding topographic effects were provided to overcome the disadvantage of planar approximations. Second, a gravimetric geoid model was constructed using the proposed StokesHelmert0 s scheme with a heterogeneous data set. Third, a least squares adjustment method combined with a multi-surface function model was employed to remove the bias between the gravimetric geoid model and the GNSS/leveling data and to refine the final local geoid model. The accuracy of the final geoid model was evaluated using independent GNSS/leveling data. Numerical results show that an external precision of 1.45 cm is achievable.展开更多
Under investigation in this paper are two coupled integrable dispersionless (CID) equations modelingthe dynamics of the current-fed string within an external magnetic field.Through a set of the dependent variabletrans...Under investigation in this paper are two coupled integrable dispersionless (CID) equations modelingthe dynamics of the current-fed string within an external magnetic field.Through a set of the dependent variabletransformations, the bilinear forms for the CID equations are derived.Based on the Hirota method and symboliccomputation, the analytic N-soliton solutions are presented.Infinitely many conservation laws for the CID equationsare given through the known spectral problem.Propagation characteristics and interaction behaviors of the solitons areanalyzed graphically.展开更多
Algebraic methods and rapid deforming techniques are used to generate three-dimensional boundary-fitted dynamic grids for assemblies. The conservative full-potential equation is solved by a time-accurate approximate f...Algebraic methods and rapid deforming techniques are used to generate three-dimensional boundary-fitted dynamic grids for assemblies. The conservative full-potential equation is solved by a time-accurate approximate factorization algorithm and internal Newton iterations. An integral boundary layer method based on the dissipation integral is used to account for viscous effects. The computational results about unsteady transonic forces on wings, bodies and control surfaces are in agreement with experimental data.展开更多
We present an overview of approaches to selfvalidating?one-dimensional integration quadrature formulas and?a verified numerical integration algorithm with an adaptive?strategy. The new interval integration adaptive al...We present an overview of approaches to selfvalidating?one-dimensional integration quadrature formulas and?a verified numerical integration algorithm with an adaptive?strategy. The new interval integration adaptive algorithm delivers?a desired integral enclosure with an error bounded by a specified?error bound. The adaptive technique is usually much more?efficient than Simpson’s rule and narrow interval results can?be reached.展开更多
Three recent breakthroughs due to AI in arts and science serve as motivation:An award winning digital image,protein folding,fast matrix multiplication.Many recent developments in artificial neural networks,particularl...Three recent breakthroughs due to AI in arts and science serve as motivation:An award winning digital image,protein folding,fast matrix multiplication.Many recent developments in artificial neural networks,particularly deep learning(DL),applied and relevant to computational mechanics(solid,fluids,finite-element technology)are reviewed in detail.Both hybrid and pure machine learning(ML)methods are discussed.Hybrid methods combine traditional PDE discretizations with ML methods either(1)to help model complex nonlinear constitutive relations,(2)to nonlinearly reduce the model order for efficient simulation(turbulence),or(3)to accelerate the simulation by predicting certain components in the traditional integration methods.Here,methods(1)and(2)relied on Long-Short-Term Memory(LSTM)architecture,with method(3)relying on convolutional neural networks.Pure ML methods to solve(nonlinear)PDEs are represented by Physics-Informed Neural network(PINN)methods,which could be combined with attention mechanism to address discontinuous solutions.Both LSTM and attention architectures,together with modern and generalized classic optimizers to include stochasticity for DL networks,are extensively reviewed.Kernel machines,including Gaussian processes,are provided to sufficient depth for more advanced works such as shallow networks with infinite width.Not only addressing experts,readers are assumed familiar with computational mechanics,but not with DL,whose concepts and applications are built up from the basics,aiming at bringing first-time learners quickly to the forefront of research.History and limitations of AI are recounted and discussed,with particular attention at pointing out misstatements or misconceptions of the classics,even in well-known references.Positioning and pointing control of a large-deformable beam is given as an example.展开更多
Simulations of contact problems involving at least one plastic solid may be costly due to their strong nonlinearity and requirements of stability.In this work,we develop an explicit asynchronous variational integrator...Simulations of contact problems involving at least one plastic solid may be costly due to their strong nonlinearity and requirements of stability.In this work,we develop an explicit asynchronous variational integrator(AVI)for inelastic non-frictional contact problems involving a plastic solid.The AVI assigns each element in the mesh an independent time step and updates the solution at the elements and nodes asynchronously.This asynchrony makes the AVI highly efficient in solving such bi-material problems.Taking advantage of the AVI,the constitutive update is locally performed in one element at a time,and contact constraints are also enforced on only one element.The time step of the contact element is subdivided into multiple segments,and the fields are updated accordingly.During a contact event,only one element involving a few degrees of freedom is considered,leading to high efficiency.The proposed formulation is first verified with a pure elastodynamics benchmark and further applied to a contact problem involving an elastoplastic solid with non-associative volumetric hardening.The numerical results indicate that the AVI exhibits excellent energy behaviors and has high computational efficiency.展开更多
1.Introduction Computational Fluid Dynamics-Discrete Element Method(CFD-DEM)is a powerful tool for simulating dense gas-solid reacting flows,which is essential in combustion,metallurgy,and waste management.Traditional...1.Introduction Computational Fluid Dynamics-Discrete Element Method(CFD-DEM)is a powerful tool for simulating dense gas-solid reacting flows,which is essential in combustion,metallurgy,and waste management.Traditional methods face challenges in CFD-DEM modeling of dense gas-solid flows due to multi-scale characteristics,limiting resolution and creating simulation bottlenecks.By integrating fluid dynamics and particle behavior,it optimizes industrial processes.This review highlights advancements,applications,and challenges,emphasizing its role in sustainable engineering.展开更多
A complete boundary integral formulation for incompressible Navier Stokes equations with time discretization by operator splitting is developed by using the fundamental solutions of the Helmhotz operator equation wit...A complete boundary integral formulation for incompressible Navier Stokes equations with time discretization by operator splitting is developed by using the fundamental solutions of the Helmhotz operator equation with different orders. The numerical results for the lift and the drag hysteresis associated with a NACA0012 aerofoil oscillating in pitch are good in comparison with available experimental data.展开更多
In this work,a consistent and physically accurate implementation of the general framework of unified second-order time accurate integrators via the well-known GSSSS framework in the Discrete Element Method is presente...In this work,a consistent and physically accurate implementation of the general framework of unified second-order time accurate integrators via the well-known GSSSS framework in the Discrete Element Method is presented.The improved tangential displacement evaluation in the present implementation of the discrete element method has been derived and implemented to preserve the consistency of the correct time level evaluation during the time integration process in calculating the algorithmic tangential displacement.Several numerical examples have been used to validate the proposed tangential displacement evaluation;this is in contrast to past practices which only seem to attain the first-order time accuracy due to inconsistent time level implementation with different algorithms for normal and tangential directions.The comparisons with the existing implementation and the superiority of the proposed implementation are given in terms of the convergence rate with improved numerical accuracy in time.Moreover,several schemes via the unified second-order time integrators within the framework of the GSSSS family have been carried out based on the proposed correct implementation.All the numerical results demonstrate that using the existing state-of-the-art implementation reduces the time accuracy to be first-order accurate in time,while the proposed implementation preserves the correct time accuracy to yield second-order.展开更多
In this paper, we investigate the Rotating N Loop-Soliton solution of the coupled integrable dispersionless equation (CIDE) that describes a current-fed string within an external magnetic field in 2D-space. Through a ...In this paper, we investigate the Rotating N Loop-Soliton solution of the coupled integrable dispersionless equation (CIDE) that describes a current-fed string within an external magnetic field in 2D-space. Through a set of independent variable transformation, we derive the bilinear form of the CIDE Equation. Based on the Hirota’s method, Perturbation technique and Symbolic computation, we present the analytic N-rotating loop soliton solution and proceed to some illustrations by presenting the cases of three- and four-soliton solutions.展开更多
In this paper two classes of equivalence transform methods for solving ordinary differential equations are proposed. One class of method is the equivalence integral transform method for special differential algebraic ...In this paper two classes of equivalence transform methods for solving ordinary differential equations are proposed. One class of method is the equivalence integral transform method for special differential algebraic problems. The advantage of this class of method is such that the amount of work calculating one integration with parameters becomes that of two interpolations, when the system of nonlinear equations is solved on the right hand side function. The other class of method is the equivalence substitution method for avoiding calculating derivative on the right hand side function. In order to avoid calculation derivatives, two equivalence substitution methods are proposed here. The application instances of some special effect of the equivalence substitution methods are given.展开更多
Tow different computer calculation methods for distortion of the wide-band diode bridge track and hold amplifier (THA) are presented based on a high frequency Schottky diode model. One of the computer programs calcula...Tow different computer calculation methods for distortion of the wide-band diode bridge track and hold amplifier (THA) are presented based on a high frequency Schottky diode model. One of the computer programs calculates the distortion of weekly nonlinear THA based on the KCL and the nonlinear-current method. The other calculates the weekly nonlinear distortion by using a Volterra series method and a nodal formulation. Comparative calculation results for the diode bridge THA have shown good agreement with these two computer program calculation methods, whereas the overall computational efficiency of the nonlinear-current method is better than that of the nodal formulation method in a special evaluation.展开更多
基金the support of the National Natural Science Foundation of China(Grant Nos.11472067 and 51609034)the Science Foundation of Liaoning Province of China(No.2021-MS-119)+1 种基金the Dalian Youth Science and Technology Star Project(No.2018RQ06)the Fundamental Research Funds for the Central Universities(Grant No.DUT20GJ216).
文摘The difficulty in solving stochastic dynamics problems lies in the need for a large number of repeated computations of deterministic dynamic equations,which has been a challenge in stochastic dynamics analysis and was discussed in this study.To efficiently and accurately compute the exponential of the dynamics state matrix and the matrix functions due to external loads,an adaptively filtered precise integration method was proposed,which inherits the high precision of the precise integrationmethod,improves the computational efficiency and saves the memory required.Moreover,the perturbation method was introduced to avoid repeated computations of matrix exponential and terms due to external loads.Based on the filtering and perturbation techniques,an adaptively filtered precise integration method considering perturbation for stochastic dynamics problems was developed.Two numerical experiments,including a model of phononic crystal and a bridge model considering random parameters,were performed to test the performance of the proposed method in terms of accuracy and efficiency.Numerical results show that the accuracy and efficiency of the proposed method are better than those of the existing precise integration method,the Newmark-βmethod and the Wilson-θmethod.
基金sponsored by the National Natural Science Foundation of China (No. 41504012)
文摘The high-precision local geoid model was computed based on the improved Stokes-Helmert0 s boundary value problem and strict integrals of topographic effects. This proposed method involves three steps.First, the mathematical form of Stokes-Helmert0 s boundary value problem was derived, and strict computational formulas regarding topographic effects were provided to overcome the disadvantage of planar approximations. Second, a gravimetric geoid model was constructed using the proposed StokesHelmert0 s scheme with a heterogeneous data set. Third, a least squares adjustment method combined with a multi-surface function model was employed to remove the bias between the gravimetric geoid model and the GNSS/leveling data and to refine the final local geoid model. The accuracy of the final geoid model was evaluated using independent GNSS/leveling data. Numerical results show that an external precision of 1.45 cm is achievable.
基金Supported by the National Natural Science Foundation of China under Grant No.60772023the Open Fund No.BUAA-SKLSDE-09KF-04+2 种基金Supported Project No.SKLSDE-2010ZX-07 of the State Key Laboratory of Software Development Environment,Beijing University of Aeronautics and Astronauticsthe National Basic Research Program of China (973 Program) under Grant No.2005CB321901 the Specialized Research Fund for the Doctoral Program of Higher Education under Grant No.200800130006,Chinese Ministry of Education
文摘Under investigation in this paper are two coupled integrable dispersionless (CID) equations modelingthe dynamics of the current-fed string within an external magnetic field.Through a set of the dependent variabletransformations, the bilinear forms for the CID equations are derived.Based on the Hirota method and symboliccomputation, the analytic N-soliton solutions are presented.Infinitely many conservation laws for the CID equationsare given through the known spectral problem.Propagation characteristics and interaction behaviors of the solitons areanalyzed graphically.
基金Aeronautical Science Foundation of China (99A52007)
文摘Algebraic methods and rapid deforming techniques are used to generate three-dimensional boundary-fitted dynamic grids for assemblies. The conservative full-potential equation is solved by a time-accurate approximate factorization algorithm and internal Newton iterations. An integral boundary layer method based on the dissipation integral is used to account for viscous effects. The computational results about unsteady transonic forces on wings, bodies and control surfaces are in agreement with experimental data.
文摘We present an overview of approaches to selfvalidating?one-dimensional integration quadrature formulas and?a verified numerical integration algorithm with an adaptive?strategy. The new interval integration adaptive algorithm delivers?a desired integral enclosure with an error bounded by a specified?error bound. The adaptive technique is usually much more?efficient than Simpson’s rule and narrow interval results can?be reached.
文摘Three recent breakthroughs due to AI in arts and science serve as motivation:An award winning digital image,protein folding,fast matrix multiplication.Many recent developments in artificial neural networks,particularly deep learning(DL),applied and relevant to computational mechanics(solid,fluids,finite-element technology)are reviewed in detail.Both hybrid and pure machine learning(ML)methods are discussed.Hybrid methods combine traditional PDE discretizations with ML methods either(1)to help model complex nonlinear constitutive relations,(2)to nonlinearly reduce the model order for efficient simulation(turbulence),or(3)to accelerate the simulation by predicting certain components in the traditional integration methods.Here,methods(1)and(2)relied on Long-Short-Term Memory(LSTM)architecture,with method(3)relying on convolutional neural networks.Pure ML methods to solve(nonlinear)PDEs are represented by Physics-Informed Neural network(PINN)methods,which could be combined with attention mechanism to address discontinuous solutions.Both LSTM and attention architectures,together with modern and generalized classic optimizers to include stochasticity for DL networks,are extensively reviewed.Kernel machines,including Gaussian processes,are provided to sufficient depth for more advanced works such as shallow networks with infinite width.Not only addressing experts,readers are assumed familiar with computational mechanics,but not with DL,whose concepts and applications are built up from the basics,aiming at bringing first-time learners quickly to the forefront of research.History and limitations of AI are recounted and discussed,with particular attention at pointing out misstatements or misconceptions of the classics,even in well-known references.Positioning and pointing control of a large-deformable beam is given as an example.
基金support from the Hui-Chun Chin and Tsung-Dao Lee Chinese Undergraduate Research Endowment(CURE).
文摘Simulations of contact problems involving at least one plastic solid may be costly due to their strong nonlinearity and requirements of stability.In this work,we develop an explicit asynchronous variational integrator(AVI)for inelastic non-frictional contact problems involving a plastic solid.The AVI assigns each element in the mesh an independent time step and updates the solution at the elements and nodes asynchronously.This asynchrony makes the AVI highly efficient in solving such bi-material problems.Taking advantage of the AVI,the constitutive update is locally performed in one element at a time,and contact constraints are also enforced on only one element.The time step of the contact element is subdivided into multiple segments,and the fields are updated accordingly.During a contact event,only one element involving a few degrees of freedom is considered,leading to high efficiency.The proposed formulation is first verified with a pure elastodynamics benchmark and further applied to a contact problem involving an elastoplastic solid with non-associative volumetric hardening.The numerical results indicate that the AVI exhibits excellent energy behaviors and has high computational efficiency.
文摘1.Introduction Computational Fluid Dynamics-Discrete Element Method(CFD-DEM)is a powerful tool for simulating dense gas-solid reacting flows,which is essential in combustion,metallurgy,and waste management.Traditional methods face challenges in CFD-DEM modeling of dense gas-solid flows due to multi-scale characteristics,limiting resolution and creating simulation bottlenecks.By integrating fluid dynamics and particle behavior,it optimizes industrial processes.This review highlights advancements,applications,and challenges,emphasizing its role in sustainable engineering.
文摘A complete boundary integral formulation for incompressible Navier Stokes equations with time discretization by operator splitting is developed by using the fundamental solutions of the Helmhotz operator equation with different orders. The numerical results for the lift and the drag hysteresis associated with a NACA0012 aerofoil oscillating in pitch are good in comparison with available experimental data.
文摘In this work,a consistent and physically accurate implementation of the general framework of unified second-order time accurate integrators via the well-known GSSSS framework in the Discrete Element Method is presented.The improved tangential displacement evaluation in the present implementation of the discrete element method has been derived and implemented to preserve the consistency of the correct time level evaluation during the time integration process in calculating the algorithmic tangential displacement.Several numerical examples have been used to validate the proposed tangential displacement evaluation;this is in contrast to past practices which only seem to attain the first-order time accuracy due to inconsistent time level implementation with different algorithms for normal and tangential directions.The comparisons with the existing implementation and the superiority of the proposed implementation are given in terms of the convergence rate with improved numerical accuracy in time.Moreover,several schemes via the unified second-order time integrators within the framework of the GSSSS family have been carried out based on the proposed correct implementation.All the numerical results demonstrate that using the existing state-of-the-art implementation reduces the time accuracy to be first-order accurate in time,while the proposed implementation preserves the correct time accuracy to yield second-order.
文摘In this paper, we investigate the Rotating N Loop-Soliton solution of the coupled integrable dispersionless equation (CIDE) that describes a current-fed string within an external magnetic field in 2D-space. Through a set of independent variable transformation, we derive the bilinear form of the CIDE Equation. Based on the Hirota’s method, Perturbation technique and Symbolic computation, we present the analytic N-rotating loop soliton solution and proceed to some illustrations by presenting the cases of three- and four-soliton solutions.
基金The project was supported by the National Natural Science Faundation of China
文摘In this paper two classes of equivalence transform methods for solving ordinary differential equations are proposed. One class of method is the equivalence integral transform method for special differential algebraic problems. The advantage of this class of method is such that the amount of work calculating one integration with parameters becomes that of two interpolations, when the system of nonlinear equations is solved on the right hand side function. The other class of method is the equivalence substitution method for avoiding calculating derivative on the right hand side function. In order to avoid calculation derivatives, two equivalence substitution methods are proposed here. The application instances of some special effect of the equivalence substitution methods are given.
文摘Tow different computer calculation methods for distortion of the wide-band diode bridge track and hold amplifier (THA) are presented based on a high frequency Schottky diode model. One of the computer programs calculates the distortion of weekly nonlinear THA based on the KCL and the nonlinear-current method. The other calculates the weekly nonlinear distortion by using a Volterra series method and a nodal formulation. Comparative calculation results for the diode bridge THA have shown good agreement with these two computer program calculation methods, whereas the overall computational efficiency of the nonlinear-current method is better than that of the nodal formulation method in a special evaluation.
