In this paper,a deep collocation method(DCM)for thin plate bending problems is proposed.This method takes advantage of computational graphs and backpropagation algorithms involved in deep learning.Besides,the proposed...In this paper,a deep collocation method(DCM)for thin plate bending problems is proposed.This method takes advantage of computational graphs and backpropagation algorithms involved in deep learning.Besides,the proposed DCM is based on a feedforward deep neural network(DNN)and differs from most previous applications of deep learning for mechanical problems.First,batches of randomly distributed collocation points are initially generated inside the domain and along the boundaries.A loss function is built with the aim that the governing partial differential equations(PDEs)of Kirchhoff plate bending problems,and the boundary/initial conditions are minimised at those collocation points.A combination of optimizers is adopted in the backpropagation process to minimize the loss function so as to obtain the optimal hyperparameters.In Kirchhoff plate bending problems,the C^1 continuity requirement poses significant difficulties in traditional mesh-based methods.This can be solved by the proposed DCM,which uses a deep neural network to approximate the continuous transversal deflection,and is proved to be suitable to the bending analysis of Kirchhoff plate of various geometries.展开更多
With extensive applications of space geodesy, three-dimensional datum transformation model has been necessarily used to transform the coordinates in the different coordinate systems.Its essence is to predict the coord...With extensive applications of space geodesy, three-dimensional datum transformation model has been necessarily used to transform the coordinates in the different coordinate systems.Its essence is to predict the coordinates of non-common points in the second coordinate system based on their coordinates in the first coordinate system and the coordinates of common points in two coordinate systems.Traditionally, the computation of seven transformation parameters and the transformation of noncommon points are individually implemented, in which the errors of coordinates are taken into account only in the second system although the coordinates in both two systems are inevitably contaminated by the random errors.Moreover, the coordinate errors of non-common points are disregarded when they are transformed using the solved transformation parameters.Here we propose the seamless (rigorous) datum transformation model to compute the transformation parameters and transform the non-common points integratively, considering the errors of all coordinates in both coordinate systems.As a result, a nonlinear coordinate transformation model is formulated.Based on the Gauss-Newton algorithm and the numerical characteristics of transformation parameters, two linear versions of the established nonlinear model are individually derived.Then the least-squares collocation (prediction) method is employed to trivially solve these linear models.Finally, the simulation experiment is carried out to demonstrate the performance and benefits of the presented method.The results show that the presented method can significantly improve the precision of the coordinate transformation, especially when the non-common points are strongly correlated with the common points used to compute the transformation parameters.展开更多
In this paper,vibration analysis of functionally graded porous beams is carried out using the third-order shear deformation theory.The beams have uniform and non-uniform porosity distributions across their thickness a...In this paper,vibration analysis of functionally graded porous beams is carried out using the third-order shear deformation theory.The beams have uniform and non-uniform porosity distributions across their thickness and both ends are supported by rotational and translational springs.The material properties of the beams such as elastic moduli and mass density can be related to the porosity and mass coefficient utilizing the typical mechanical features of open-cell metal foams.The Chebyshev collocation method is applied to solve the governing equations derived from Hamilton's principle,which is used in order to obtain the accurate natural frequencies for the vibration problem of beams with various general and elastic boundary conditions.Based on the numerical experiments,it is revealed that the natural frequencies of the beams with asymmetric and non-uniform porosity distributions are higher than those of other beams with uniform and symmetric porosity distributions.展开更多
A combined deep machine learning(DML)and collocation based approach to solve the partial differential equations using artificial neural networks is proposed.The developed method is applied to solve problems governed b...A combined deep machine learning(DML)and collocation based approach to solve the partial differential equations using artificial neural networks is proposed.The developed method is applied to solve problems governed by the Sine–Gordon equation(SGE),the scalar wave equation and elasto-dynamics.Two methods are studied:one is a space-time formulation and the other is a semi-discrete method based on an implicit Runge–Kutta(RK)time integration.The methodology is implemented using the Tensorflow framework and it is tested on several numerical examples.Based on the results,the relative normalized error was observed to be less than 5%in all cases.展开更多
The collocation method is a widely used numerical method for science and engineering problems governed by partial differential equations.This paper provides a comprehensive review of collocation methods and their appl...The collocation method is a widely used numerical method for science and engineering problems governed by partial differential equations.This paper provides a comprehensive review of collocation methods and their applications,focused on elasticity,heat conduction,electromagnetic field analysis,and fluid dynamics.The merits of the collocation method can be attributed to the need for element mesh,simple implementation,high computational efficiency,and ease in handling irregular domain problems since the collocation method is a type of node-based numerical method.Beginning with the fundamental principles of the collocation method,the discretization process in the continuous domain is elucidated,and how the collocation method approximation solutions for solving differential equations are explained.Delving into the historical development of the collocation methods,their earliest applications and key milestones are traced,thereby demonstrating their evolution within the realm of numerical computation.The mathematical foundations of collocation methods,encompassing the selection of interpolation functions,definition of weighting functions,and derivation of integration rules,are examined in detail,emphasizing their significance in comprehending the method’s effectiveness and stability.At last,the practical application of the collocation methods in engineering contexts is emphasized,including heat conduction simulations,electromagnetic coupled field analysis,and fluid dynamics simulations.These specific case studies can underscore collocation method’s broad applicability and effectiveness in addressing complex engineering challenges.In conclusion,this paper puts forward the future development trend of the collocation method through rigorous analysis and discussion,thereby facilitating further advancements in research and practical applications within these fields.展开更多
Meshless method is a type of promising numerical approach. But for the method, the convergence is still lack of common theoretical explanations, and the technique of numerical implementation also remains to be improve...Meshless method is a type of promising numerical approach. But for the method, the convergence is still lack of common theoretical explanations, and the technique of numerical implementation also remains to be improved. It is worth noting that a kind of uniformly defined intervention point is used in many existing schemes. Therefore, the intervention-point principle is proposed. The viewpoint is likely to give a reasonable explanation for the inaccuracy and instability of the collocation method. Based on the principle, a design process for a new scheme was demonstrated. Some initial numerical tests were also offered. The results have revealed the intervention point to take effect on convergence, suggested a construction concept using intervention point for meshless collocation method, and presented a new scheme of meshless method for application.展开更多
文摘In this paper,a deep collocation method(DCM)for thin plate bending problems is proposed.This method takes advantage of computational graphs and backpropagation algorithms involved in deep learning.Besides,the proposed DCM is based on a feedforward deep neural network(DNN)and differs from most previous applications of deep learning for mechanical problems.First,batches of randomly distributed collocation points are initially generated inside the domain and along the boundaries.A loss function is built with the aim that the governing partial differential equations(PDEs)of Kirchhoff plate bending problems,and the boundary/initial conditions are minimised at those collocation points.A combination of optimizers is adopted in the backpropagation process to minimize the loss function so as to obtain the optimal hyperparameters.In Kirchhoff plate bending problems,the C^1 continuity requirement poses significant difficulties in traditional mesh-based methods.This can be solved by the proposed DCM,which uses a deep neural network to approximate the continuous transversal deflection,and is proved to be suitable to the bending analysis of Kirchhoff plate of various geometries.
基金supported by National Basic Research Program of China(Grant No.2012CB957703)the National Natural Science Foundation of China(Grant Nos.41074018 and 41104002)
文摘With extensive applications of space geodesy, three-dimensional datum transformation model has been necessarily used to transform the coordinates in the different coordinate systems.Its essence is to predict the coordinates of non-common points in the second coordinate system based on their coordinates in the first coordinate system and the coordinates of common points in two coordinate systems.Traditionally, the computation of seven transformation parameters and the transformation of noncommon points are individually implemented, in which the errors of coordinates are taken into account only in the second system although the coordinates in both two systems are inevitably contaminated by the random errors.Moreover, the coordinate errors of non-common points are disregarded when they are transformed using the solved transformation parameters.Here we propose the seamless (rigorous) datum transformation model to compute the transformation parameters and transform the non-common points integratively, considering the errors of all coordinates in both coordinate systems.As a result, a nonlinear coordinate transformation model is formulated.Based on the Gauss-Newton algorithm and the numerical characteristics of transformation parameters, two linear versions of the established nonlinear model are individually derived.Then the least-squares collocation (prediction) method is employed to trivially solve these linear models.Finally, the simulation experiment is carried out to demonstrate the performance and benefits of the presented method.The results show that the presented method can significantly improve the precision of the coordinate transformation, especially when the non-common points are strongly correlated with the common points used to compute the transformation parameters.
文摘In this paper,vibration analysis of functionally graded porous beams is carried out using the third-order shear deformation theory.The beams have uniform and non-uniform porosity distributions across their thickness and both ends are supported by rotational and translational springs.The material properties of the beams such as elastic moduli and mass density can be related to the porosity and mass coefficient utilizing the typical mechanical features of open-cell metal foams.The Chebyshev collocation method is applied to solve the governing equations derived from Hamilton's principle,which is used in order to obtain the accurate natural frequencies for the vibration problem of beams with various general and elastic boundary conditions.Based on the numerical experiments,it is revealed that the natural frequencies of the beams with asymmetric and non-uniform porosity distributions are higher than those of other beams with uniform and symmetric porosity distributions.
基金the funds from the Department of Science and Technology(DST),Science and Engineering Research Board(SERB),India(No.SRG/2019/001581).
文摘A combined deep machine learning(DML)and collocation based approach to solve the partial differential equations using artificial neural networks is proposed.The developed method is applied to solve problems governed by the Sine–Gordon equation(SGE),the scalar wave equation and elasto-dynamics.Two methods are studied:one is a space-time formulation and the other is a semi-discrete method based on an implicit Runge–Kutta(RK)time integration.The methodology is implemented using the Tensorflow framework and it is tested on several numerical examples.Based on the results,the relative normalized error was observed to be less than 5%in all cases.
基金the National Natural Science Foundation of China for financial support to this work under Grant NSFC No.12072064.
文摘The collocation method is a widely used numerical method for science and engineering problems governed by partial differential equations.This paper provides a comprehensive review of collocation methods and their applications,focused on elasticity,heat conduction,electromagnetic field analysis,and fluid dynamics.The merits of the collocation method can be attributed to the need for element mesh,simple implementation,high computational efficiency,and ease in handling irregular domain problems since the collocation method is a type of node-based numerical method.Beginning with the fundamental principles of the collocation method,the discretization process in the continuous domain is elucidated,and how the collocation method approximation solutions for solving differential equations are explained.Delving into the historical development of the collocation methods,their earliest applications and key milestones are traced,thereby demonstrating their evolution within the realm of numerical computation.The mathematical foundations of collocation methods,encompassing the selection of interpolation functions,definition of weighting functions,and derivation of integration rules,are examined in detail,emphasizing their significance in comprehending the method’s effectiveness and stability.At last,the practical application of the collocation methods in engineering contexts is emphasized,including heat conduction simulations,electromagnetic coupled field analysis,and fluid dynamics simulations.These specific case studies can underscore collocation method’s broad applicability and effectiveness in addressing complex engineering challenges.In conclusion,this paper puts forward the future development trend of the collocation method through rigorous analysis and discussion,thereby facilitating further advancements in research and practical applications within these fields.
基金supported by the National Natural Science Foundation of China(51038002)the Planned Science and Technology Project of Hunan Province, China (2010FJ6023)the Open Fund of Key Laboratory of Road Structure and Material of Ministry of Transport (Changsha, kfj100201)
文摘Meshless method is a type of promising numerical approach. But for the method, the convergence is still lack of common theoretical explanations, and the technique of numerical implementation also remains to be improved. It is worth noting that a kind of uniformly defined intervention point is used in many existing schemes. Therefore, the intervention-point principle is proposed. The viewpoint is likely to give a reasonable explanation for the inaccuracy and instability of the collocation method. Based on the principle, a design process for a new scheme was demonstrated. Some initial numerical tests were also offered. The results have revealed the intervention point to take effect on convergence, suggested a construction concept using intervention point for meshless collocation method, and presented a new scheme of meshless method for application.
