Multi-objective Bayesian topology optimization of a lattice-structured heat sink in natural convection

Abstract Additive manufacturing (AM) has an affinity with topology optimization to think of various designs complex structures. Hence, this paper aims optimize the design a lattice-structured heat sink, which can be manufactured by AM. The objectives are maximize thermal performance convective transfer in natural convection simulated computational fluid dynamics (CFD) and minimize material cost required for AM process at same time. lattice structure is represented as node/edge system via graph theory moderate number variables. Bayesian optimization, employs non-dominated sorting genetic algorithm II Kriging surrogate model, conducted search better minimum CFD cost. present successfully finds sink than reference fin-structured regarding Also, several optimized outperform pin-fin-structured though pin-fin still advantageous cost-oriented design. This also discusses flow mechanism observed explain how satisfies competing simultaneously.