Comsol Platform-based Continuum Structural Topology Optimization Method

As a new research field the of structural design, topology optimization could have a better optimization efficiency due to fact that it can give the designer a more innovation idea in the conceptual design phase of the structural design, therefore gained a lot attention from the researcher and became a hot topic in the structural design research field. Base on the review of the domestic and foreign research, this paper presents a way to using COMSOL with MATLAB package to implement the topology optimization method. The main content including:(1)Base on the SIMP (Solid Isotropic Material with Penalization) model, present a way to solve the topology optimization problem with the goal of minimizes the compliance under the volume constraint. Illustrate how to use the MMA (The Method of Moving Asymptotes) algorithm to solve this mathematical model. Develop a way to using COMSOL with MATLAB package to implement this algorithm. At last, several examples were showed to illustrate its effectiveness to solve the topology optimization problems.(2)Presents a way to solve the topology optimization problem with stress constraint, using a macro stress constraint as the stress constrain condition, it can avoid the stress singularity problems caused by the local stress constraint. Also using COMSOL with MATLAB package to implement this algorithm and several examples showed that it can solve the topology optimization problem with the goal of obtain the design with minimized compliance under the stress constraint using the most less volume of material.(3)Develop a way to solve the reliability based topology optimization problem. Using the SORA (Sequential Optimization and Reliability Assessment) process, first solve the deterministic topology optimization problem based on the SIMP model, then use the Response Surface method combining with RIA(Reliability Index Approach) to assess the reliability, at last check the convergence of the whole process using the reliability constraint or update the design variables for the new iteration until the whole problem reach the optimum value. Two examples are given to prove the effectiveness of this method, and Monte Carlo simulations are carried out to verify its correctness.