COMPLEMENTARY TASKS IN THE NUMERICAL OPTIMIZATION OF A COOLING DEVICE THROUGH THE PARAMETRIC STUDY AND USING A GENERATIVE DESIGN TOOL

This paper presents the results of the numerical optimization of a cooling device obtained by means of the standard parametric study as well as the advanced generative design tool based on the adjoint optimization method. The thermal analysis of the cooling device included the simulation of the conjugate heat transfer (taking the example from the power electronics), comprising the heat generation within the solid components attached to the device body, the heat conduction through the device body with coolant channel, and the convective heat removal by the coolant flow. In the first step, the standard parametric study was performed using the open-source CFD library suite OpenFOAM to determine the main geometry features of the coolant channel with pins. In the next step, the generative design tool ColdStream was used to refine the channel pre-design through the topology optimization for the specified objective function. In the present case, the solution was sought for the minimization of the overall pressure drop, in addition to the even temperature distribution across the heated components. The paper describes the workflow of the complementary optimization procedure, presenting the outcomes of the numerical analysis with the deployed approaches, and in conclusion discussing the strengths and weaknesses of both approaches.