Topology optimization for local thickness constraint by using virtual phisical model

Abstract

Although it is important to apply thickness constraints to topology optimization problems, it is generally difficult to provide geometric features such as thickness as constraints. In this study, we attempted to formulate topology optimization with local thicknesses by extracting geometric features using partial differential equations and defining new thicknesses from the state field of each point in the object domain. The topological derivative is obtained by setting the constraint function so that the defined local thickness is constant and calculating the associated field based on the adjoint variable method. We conducted a basic study to apply the local thickness constraint to the topological optimization problem, by using this topological derivative for shape updating.