Optimal design of a ram air cooling duct housing the condenser of an airborne ORC WHR unit

Thermal energy harvesting from aero engines using compact organic Rankine cycle (ORC) units can reduce aircraft fuel consumption and emissions. Heat exchangers (HX) are key components of such ORC systems: the primary heat exchanger, or evaporator, recovers part of the thermal energy of the exhaust gases of the engine, while the condenser rejects the thermal energy required to condense the working fluid into the ambient through ram air. The use of ram air can significantly increase the overall drag of the aircraft, thus reducing the fuel savings achievable through the ORC system. As a result, the proper design of the ram air duct and the condenser is critical for the feasibility of thermal energy harvesting from aero engines. By exploiting the ram air temperature increase across the heat exchanger and minimizing pressure losses in the duct, it is possible to reduce the ram air drag or even generate net thrust. This possibility is strongly related to the system design and constraints, thus the integration of fast yet accurate models for the preliminary design of the ram air duct with the overall system model of the combined cycle engine is key. The literature currently lacks studies that assess the accuracy of ram air duct lumped parameter models or provide guidelines for the design of air ducts housing HXs. This paper thus focuses on the development of a quasi-1D model for predicting the drag associated with a ram air duct housing the condenser of a turboshaft ORC bottoming unit and its verification against the results of CFD simulations in which the HX is modeled as a porous medium. The quasi-1D model is then used to perform a sensitivity study and a series of constrained optimizations in which the maximum allowed duct length is varied. More specifically, the paper first describes the lumped parameter model of the duct used to assess the influence of heat exchanger geometry and key design parameters of the duct on the performance of the ram air cooling system. Then, a constrained optimization is carried out to identify the optimal design solution for the duct. Finally, a CFD model of the duct is developed by resorting to the porous media modeling methodology to simulate the presence of the condenser. The porous zone source terms are calibrated using in-house heat exchanger design and rating tools. Results demonstrate that the space constraints strongly affect the design of the ram air duct and that the condenser is one of the main pressure loss sources. Additionally, it is shown that in a preliminary design stage, the lumped parameter model can predict the internal drag of the duct with reasonable accuracy.