Speaker
Description
The thermal design of equipment and space platforms is a difficult task due to the large variability of the thermal conditions during the orbit and the strict constraints imposed by other subsystems. In the past years, several strategies have been studied to meet these constraints while reducing the signature of the thermal control subsystem. These techniques encompass different algorithms that give the optimal placements of the components in the platform. When this relocation is not possible, another alternative is to modify the thermal couplings between the components, obtaining a more efficient heat distribution. To achieve this, topology optimization, originally linked to lightweight structure optimization, has been extended to heat transfer problems. Nevertheless, the influence of thermal radiation has been rarely taken into account in this context. In addition, most formulations are based on a finite element discretization, whereas most commercial software for heat transfer problems use the Lumped Parameter Method. This paper aims to extend the topology optimization framework to overcome the aforementioned issues. An optimization algorithm based on the Heaviside Projection Method is developed to optimize the conductive paths of two-dimensional space structures. The performance of the algorithm has been studied by optimizing a multi-case objective using a Printed Circuit Board. The method is used to generate the optimal shape of an additional copper layer so that the components on the board satisfy specific thermal requirements simultaneously for a hot and cold case. Results show a significant improvement in the thermal compliance of all the components with the addition of the high conductive layer. Moreover, a reduction method is introduced to transfer the material distribution from the optimization to a coarser mesh of any size, which is useful to facilitate its implementation in existing software, where a large number of degrees of freedom can be a limitation.
