Novel Topology Optimization Strategy Utilizing Multi-stage Acceleration Method

Topology optimization is a widely used technique for obtaining optimal material distribution. The material distribution usually must have a minimum (maximum) objective function and fulfil constraint functions. Optimization requires tuning of parameters to get optimal shape. The process is a time-consuming task due to the need for multiple runs. Neglecting the process might lead to “false-optimal” results. This paper aims to present a novel strategy using a multi-stage acceleration method. Novelty is presented in refining the initial guess of the following stage using results from the previous one. The multi-stage method is presented in three ways; Transferring between coarser mesh to fine one, restarting using a new filter, or restarting continuation. The usability of the strategy is presented using practical optimization of spring-loaded camming devices, especially its cams. Topology optimization is performed using the software MATLAB.