Multidomain Topology Optimization for Structural and Material Designs
Zheng-Dong Ma, Noboru Kikuchi, Christophe Pierre, Basavaraju Raju
Abstract
Zheng-Dong Ma, Noboru Kikuchi, Christophe Pierre, Basavaraju Raju
Abstract
A multidomain topology optimization technique (MDTO) is developed, which extends the standard topology optimization method to the realm of more realistic engineering design problems. The new technique enables the effective design of a complex engineering structure by allowing the designer to control the material distribution among the subdomains during the optimal design process, to use multiple materials or composite materials in the various subdomains of the structure, and to follow a desired pattern or tendency for the material distribution. A new algorithm of Sequential Approximate Optimization (SAO) is proposed for the multidomain topology optimization, which is an enhancement and a generalization of previous SAO algorithms (including Optimality Criteria and Convex Linearization methods, etc.). An advanced substructuring method using quasi-static modes is also introduced to condense the nodal variables associated with the multidomain topology optimization problem, especially for the nondesign subdomains. The effectiveness of the new MDTO approach is demonstrated for various design problems, including one of “structure-fixture simultaneous design,” one of “functionally graded material design,” and one of “crush energy management.” These case studies demonstrate the potential significance of the new capability developed for a wide range of engineering design problems.
OpenAlex reports 42 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
A multidomain topology optimization technique (MDTO) is developed, which extends the standard topology optimization method to the realm of more realistic engineering design problems. The new technique enables the effective design of a complex engineering structure by allowing the designer to control the material distribution among the subdomains during the optimal design process, to use multiple materials or composite materials in the various subdomains of the structure, and to follow a desired pattern or tendency for the material distribution. A new algorithm of Sequential Approximate Optimization (SAO) is proposed for the multidomain topology optimization, which is an enhancement and a generalization of previous SAO algorithms (including Optimality Criteria and Convex Linearization methods, etc.). An advanced substructuring method using quasi-static modes is also introduced to condense the nodal variables associated with the multidomain topology optimization problem, especially for the nondesign subdomains. The effectiveness of the new MDTO approach is demonstrated for various design problems, including one of “structure-fixture simultaneous design,” one of “functionally graded material design,” and one of “crush energy management.” These case studies demonstrate the potential significance of the new capability developed for a wide range of engineering design problems.
Key concepts: Topology optimization, Computer science, Engineering design process, Mathematical optimization, Topology (electrical circuits), Compliant mechanism, Engineering optimization, Optimization problem