The structure optimization design under thermo-mechanical coupling is a difficult problem in the topology optimization field.An adaptive growth algorithm has become a more effective approach for structural topology op...The structure optimization design under thermo-mechanical coupling is a difficult problem in the topology optimization field.An adaptive growth algorithm has become a more effective approach for structural topology optimization.This paper proposed a topology optimization method by an adaptive growth algorithm for the stiffener layout design of box type load-bearing components under thermo-mechanical coupling.Based on the stiffness diffusion theory,both the load stiffness matrix and the heat conduction stiffness matrix of the stiffener are spread at the same time to make sure the stiffener grows freely and obtain an optimal stiffener layout design.Meanwhile,the objectives of optimization are the minimization of strain energy and thermal compliance of the whole structure,and thermo-mechanical coupling is considered.Numerical studies for square shells clearly show the effectiveness of the proposed method for stiffener layout optimization under thermo-mechanical coupling.Finally,the method is applied to optimize the stiffener layout of box type load-bearing component of themachining center.The optimization results show that both the structural deformation and temperature of the load-bearing component with the growth stiffener layout,which are optimized by the adaptive growth algorithm,are less than the stiffener layout of shape‘#’stiffener layout.It provides a new solution approach for stiffener layout optimization design of box type load-bearing components under thermo-mechanical coupling.展开更多
Non-negative matrix factorization （NMF） is a popular feature encoding method for image understanding due to its non-negative properties in representation, but the learnt basis images are not always local due to the ...Non-negative matrix factorization （NMF） is a popular feature encoding method for image understanding due to its non-negative properties in representation, but the learnt basis images are not always local due to the lack of explicit constraints in its objective. Various algebraic or geometric local constraints are hence proposed to shape the behaviour of the original NMF. Such constraints are usually rigid in the sense that they have to be specified beforehand instead of learning from the data. In this paper, we propose a flexible spatial constraint method for NMF learning based on factor analysis. Particularly, to learn the local spatial structure of the images, we apply a series of transformations such as orthogonal rotation and thresholding to the factor loading matrix obtained through factor analysis. Then we map the transformed loading matrix into a Laplacian matrix and incorporate this into a max-margin non-negative matrix factorization framework as a penalty term, aiming to learn a representation space which is non-negative, discriminative and localstructure-preserving. We verify the feasibility and effectiveness of the proposed method on several real world datasets with encouraging results.展开更多
The panel-type structures used in aerospace engineering can be subjected to severe highfrequency acoustic loadings in service. This paper evaluates the frequency-dependent random fatigue of panel-type structures made ...The panel-type structures used in aerospace engineering can be subjected to severe highfrequency acoustic loadings in service. This paper evaluates the frequency-dependent random fatigue of panel-type structures made of ceramic matrix composites（CMCs） under acoustic loadings. Firstly, the high-frequency random responses from the broadband random excitation will result in more stress cycles in a deinite period of time. The probability density distributions of stress amplitudes will be different in different frequency bandwidths, though the peak stress estimations are identical. Secondly, the fatigue properties of CMCs can be highly frequency-dependent. The fatigue evaluation method for the random vibration case is adopted to evaluate the fatigue damage of a representative stiffened panel structure. The frequency effect through S-N curves on random fatigue damage is numerically veriied. Finally, a parameter is demonstrated to characterize the mean vibration frequency of a random process, and hence this parameter can further be considered as a reasonable loading frequency in the fatigue tests of CMCs to obtain more reliable S-N curves.Therefore, the inluence of vibration frequency can be incorporated in the random fatigue model from the two perspectives.展开更多
A cylindrical system of vector functions, the stiffness matrix method and the corresponding recursive algorithm are proposed to investigate the static response of transversely isotropic,layered magneto-electro-elastic...A cylindrical system of vector functions, the stiffness matrix method and the corresponding recursive algorithm are proposed to investigate the static response of transversely isotropic,layered magneto-electro-elastic（MEE） structures over a homogeneous half-space substrate subjected to circular surface loading. In terms of the system of vector functions, we expand the extended displacements and stresses, and deduce two sets of ordinary differential equations, which are related to the expansion coeficients. The solution to one of the two sets of these ordinary differential equations can be evaluated by using the stiffness matrix method and the corresponding recursive algorithm. These expansion coeficients are then integrated by adaptive Gaussian quadrature to obtain the displacements and stresses in the physical domain. Two types of surface loads, mechanical pressure and electric loading,are considered in the numerical examples. The calculated results show that the proposed technique is stable and effective in analyzing the layered half-space MEE structures under surface loading.展开更多
基金supported by National Natural Science Foundation of China (No.52075445)Science,Technology and Innovation Commission of Shenzhen Municipality (No.JCYJ20190806151013025).
文摘The structure optimization design under thermo-mechanical coupling is a difficult problem in the topology optimization field.An adaptive growth algorithm has become a more effective approach for structural topology optimization.This paper proposed a topology optimization method by an adaptive growth algorithm for the stiffener layout design of box type load-bearing components under thermo-mechanical coupling.Based on the stiffness diffusion theory,both the load stiffness matrix and the heat conduction stiffness matrix of the stiffener are spread at the same time to make sure the stiffener grows freely and obtain an optimal stiffener layout design.Meanwhile,the objectives of optimization are the minimization of strain energy and thermal compliance of the whole structure,and thermo-mechanical coupling is considered.Numerical studies for square shells clearly show the effectiveness of the proposed method for stiffener layout optimization under thermo-mechanical coupling.Finally,the method is applied to optimize the stiffener layout of box type load-bearing component of themachining center.The optimization results show that both the structural deformation and temperature of the load-bearing component with the growth stiffener layout,which are optimized by the adaptive growth algorithm,are less than the stiffener layout of shape‘#’stiffener layout.It provides a new solution approach for stiffener layout optimization design of box type load-bearing components under thermo-mechanical coupling.
文摘Non-negative matrix factorization （NMF） is a popular feature encoding method for image understanding due to its non-negative properties in representation, but the learnt basis images are not always local due to the lack of explicit constraints in its objective. Various algebraic or geometric local constraints are hence proposed to shape the behaviour of the original NMF. Such constraints are usually rigid in the sense that they have to be specified beforehand instead of learning from the data. In this paper, we propose a flexible spatial constraint method for NMF learning based on factor analysis. Particularly, to learn the local spatial structure of the images, we apply a series of transformations such as orthogonal rotation and thresholding to the factor loading matrix obtained through factor analysis. Then we map the transformed loading matrix into a Laplacian matrix and incorporate this into a max-margin non-negative matrix factorization framework as a penalty term, aiming to learn a representation space which is non-negative, discriminative and localstructure-preserving. We verify the feasibility and effectiveness of the proposed method on several real world datasets with encouraging results.
基金supports from the National Natural Science Foundation of China (No. 11572086 , No. 11402052 )the New Century Excellent Talent in University (NCET-11-0086)+3 种基金the Natural Science Foundation of Jiangsu province (No. BK20140616 )the Fundamental Research Funds for the Central Universities and the Scientiic Research Innovation Program of Jiangsu Province College Postgraduates (KYLX_0093, KYLX15_0092)the China Scholarship Council ( 201506090047 )the Ministry of Education, Science and Technological Development of Republic of Serbia ( TR 35011 and ON 74001 )
文摘The panel-type structures used in aerospace engineering can be subjected to severe highfrequency acoustic loadings in service. This paper evaluates the frequency-dependent random fatigue of panel-type structures made of ceramic matrix composites（CMCs） under acoustic loadings. Firstly, the high-frequency random responses from the broadband random excitation will result in more stress cycles in a deinite period of time. The probability density distributions of stress amplitudes will be different in different frequency bandwidths, though the peak stress estimations are identical. Secondly, the fatigue properties of CMCs can be highly frequency-dependent. The fatigue evaluation method for the random vibration case is adopted to evaluate the fatigue damage of a representative stiffened panel structure. The frequency effect through S-N curves on random fatigue damage is numerically veriied. Finally, a parameter is demonstrated to characterize the mean vibration frequency of a random process, and hence this parameter can further be considered as a reasonable loading frequency in the fatigue tests of CMCs to obtain more reliable S-N curves.Therefore, the inluence of vibration frequency can be incorporated in the random fatigue model from the two perspectives.
基金supported by National Natural Science Foundation of China (Nos. U1333201, 11502123 and 11262012 )
文摘A cylindrical system of vector functions, the stiffness matrix method and the corresponding recursive algorithm are proposed to investigate the static response of transversely isotropic,layered magneto-electro-elastic（MEE） structures over a homogeneous half-space substrate subjected to circular surface loading. In terms of the system of vector functions, we expand the extended displacements and stresses, and deduce two sets of ordinary differential equations, which are related to the expansion coeficients. The solution to one of the two sets of these ordinary differential equations can be evaluated by using the stiffness matrix method and the corresponding recursive algorithm. These expansion coeficients are then integrated by adaptive Gaussian quadrature to obtain the displacements and stresses in the physical domain. Two types of surface loads, mechanical pressure and electric loading,are considered in the numerical examples. The calculated results show that the proposed technique is stable and effective in analyzing the layered half-space MEE structures under surface loading.
