Search

A simple and effective filtering method to control the member size of an optimized structure is proposed for topology optimization. In the present approach, the original objective sensitivities are replaced with their relative values evaluated within a filtering area. By adjusting the size of the filtering area, the member size of an optimized structure or the level of its topological complexity can be controlled even within a given finite element mesh. In contrast to the checkerboard-free filter, the present filter focuses on high-frequency components of the sensitivities. Since the present filtering method does not add a penalty term to the objective function nor require additional constraints, it is not only efficient but also simple to implement. Mean compliance minimization and eigenfrequency maximization problems are considered to verify the effectiveness of the present approach.

The objective of the present study is to show that the numerical instability characterized by checkerboard patterns can be completely controlled when non-conforming four-node finite elements are employed. Since the convergence of the non-conforming finite element is independent of the Lame parameters, the stiffness of the non-conforming element exhibits correct limiting behaviour, which is desirable in prohibiting the unwanted formation of checkerboards in topology optimization. We employ the homogenization method to show the checkerboard-free property of the non-conforming element in topology optimization problems and verify it with three typical optimization examples.