An approach to systematically reduce the extent of the design space in topology optimization for heat transfer problems

In this paper, we present an approach how to reduce the spatial extent of the design
space in topology optimization in order to reduce computational costs. We consider heat
transfer problems of complex systems with cooling through natural convection. The
proposed approach consists of two steps, which shift the compromise between
computational cost and model quality to either side, respectively. The combination results
in a smaller design space and therefore reduced computational cost. We apply the method
to the optimization of an electro hydrostatic actuator. The proposed approach can help
the product developer in finding optimal solutions for the thermal optimization of complex
systems.