Adjoint-based shape optimization of a plate-fin heat exchanger using CFD

Plate-fin heat exchangers are common in aerospace applications because of their compact size and lightweight. These heat exchangers are critical for aircraft applications such as engine-oil cooling, environmental control systems, and thermal management of electric components. With the rise in aircraft electrification, the cooling of electric components has become increasingly crucial. Additionally, hydrogen-powered propulsion introduces several thermal management issues that require heat exchangers. By minimizing the weight and drag added by plate-fin heat exchangers, we can increase the overall efficiency of current and future aircraft. However, the analytic equations used currently to size plate-fin heat exchangers lack the fidelity required to design novel shape configurations. To address this issue, we apply CFD-based adjoint-based shape optimization to plate-fin heat exchangers to minimize drag and mass. The designs optimized for minimum drag have fin geometries stretched in either the height or width direction. We found two local minima in the design space for drag- minimization corresponding to these two stretching directions. In contrast, the mass-minimizing designs are short and use waves along the length of the channel to induce separation and promote mixing. These results highlight the potential of gradient-based shape optimization methods to generate improved heat exchangers for drag- or weight-critical aerospace applications. Furthermore, by implementing additional constraints and objective functions, this approach could be adapted to other application areas, such as power generation, automotive, air-conditioning, and industrial cryogenics.