Multi-scale and multi-material topology optimization of channel-cooling cellular structures for thermomechanical behaviors

Abstract

This paper presents a concurrent multi-scale and multi-material topology optimization approach to design cellular structures containing cooling channels for efficient thermal shielding and load carrying capabilities. The structure is composed of cooling channels, voids and metal lattices that are infilled with thermal insolation materials. A coupled thermofluidic and mechanical model with a design-dependent thermal convection is considered at the macro design scale, while the numerical homogenization and surrogate model based approaches are employed to characterize the effective thermoelastic properties of predefined quasi-periodic structure patterns at the sub-level scale. Besides, a multi-material interpolation scheme based on the Porous Anisotropic Material with Penalization (PAMP) is used for topology optimization. The proposed approach can be applied to design the multi-scale topology of the lightweight load-bearing structures, of which the thermal shielding and dissipation are key performance indices in addition to the mechanical properties. The optimized designs exhibit better thermomechanical behaviors compared to those from the mono-scale solid based counterpart at a similar computational cost because of the larger design freedom of the multi-scale composite configuration. Several design examples and verification results are given to demonstrate the applicability of the approach.