XFEM level set-based topology optimization for turbulent conjugate heat transfer problems

Abstract Solving conjugate heat transfer design problems is relevant for various engineering applications requiring efficient thermal management. Heat exchange between fluid and solid can be enhanced by optimizing the system layout shape of flow channels. As transferred at fluid/solid interfaces, it crucial to accurately resolve geometry physics responses across these interfaces. To address this challenge, work investigates first time use an eXtended Finite Element Method (XFEM) approach predict physical considering turbulent flow. This analysis integrated into a level set-based optimization framework. The domain immersed background mesh interfaces defined implicitly one or multiple set functions. functions are discretized higher-order B-splines. predicted Reynolds Averaged Navier–Stokes equations. Turbulence described Spalart–Allmaras model energy transport advection–diffusion model. element approximations augmented generalized Heaviside enrichment strategy with state fields being approximated linear basis Boundary interface conditions enforced weakly Nitsche’s method, face-oriented ghost stabilization used mitigate numerical instabilities associated emergence small integration subdomains. proposed XFEM validated against benchmark problems. Optimization solved gradient-based algorithms required sensitivity performed adjoint method. framework illustrated exchangers in two dimensions. results show that, tuning generate turbulence within exchanger, increased.