427 High-efficient Multiscale Analysis and Design for Heterogeneous Materials
Gengdong Cheng, Dalian University of Technology
Xikui Li, Dalian University of Technology
The heterogeneous materials such as fiber-reinforced or laminated composites, granular media, shape-memory alloys, biomaterials and random materials possess highly complex meso (micro/nano)-structure. As failure phenomena of heterogeneous materials are concerned, it is required in multiscale analysis and design to determine non-linear effective properties and to characterize material degradations such as stiffness and strength reduction of macroscopic continuum in terms of meso-structured material parameters and meso-scopic dissipative mechanisms. Moreover, date-driven material designs in material genome need the construction of date base.
Among different schemes proposed to deal with multiscale analysis and design for heterogeneous materials is a promising approach using the multiscale computational homogenization method, taking into account statistical nature of meso-structures to some extent. In the computational homogenization methods high-fidelity numerical simulations of meso-structured representative unit cell (RUC) models are to be performed. The scale in the number of unknowns for the problems to be solved by computational multiscale approach is usually tremendous. The development of high-efficient and accurate methods to reduce the complexity of the refined numerical modeling of both RUC and structure, then to reduce computational cost is expected. Many different schemes referred as reduced order models have been proposed. Among them is the model reduction approach proposed by Liu WK et al [1], which is attractive to deal with the burden of large computational cost of concurrent computational multiscale methods performed by direct numerical simulation.
The proposed mini-symposium aims to discuss the recent advances in developing high-efficient and accurate multiscale analysis and design methods for heterogeneous materials. The topics covered in the mini-symposium include different high efficient multi-scale schemes for analysis and design of heterogeneous materials and structures.
Reference
[1] Zeliang Liu, M.A. Bessa Wing Kam Liu. Self-consistent clustering analysis: An efficient multi-scale scheme for inelastic heterogeneous materials. Comput. Methods Appl. Mech. Engrg. 306 (2016) 319–341.
Among different schemes proposed to deal with multiscale analysis and design for heterogeneous materials is a promising approach using the multiscale computational homogenization method, taking into account statistical nature of meso-structures to some extent. In the computational homogenization methods high-fidelity numerical simulations of meso-structured representative unit cell (RUC) models are to be performed. The scale in the number of unknowns for the problems to be solved by computational multiscale approach is usually tremendous. The development of high-efficient and accurate methods to reduce the complexity of the refined numerical modeling of both RUC and structure, then to reduce computational cost is expected. Many different schemes referred as reduced order models have been proposed. Among them is the model reduction approach proposed by Liu WK et al [1], which is attractive to deal with the burden of large computational cost of concurrent computational multiscale methods performed by direct numerical simulation.
The proposed mini-symposium aims to discuss the recent advances in developing high-efficient and accurate multiscale analysis and design methods for heterogeneous materials. The topics covered in the mini-symposium include different high efficient multi-scale schemes for analysis and design of heterogeneous materials and structures.
Reference
[1] Zeliang Liu, M.A. Bessa Wing Kam Liu. Self-consistent clustering analysis: An efficient multi-scale scheme for inelastic heterogeneous materials. Comput. Methods Appl. Mech. Engrg. 306 (2016) 319–341.
