The Coupling of a Preliminary Design Method with CFD Analysis for the Design of Heat Exchangers in Aviation

A revolutionary and innovative propulsion concept has been presented by MTU Aero Engines: The Water-Enhanced Turbofan (WET). This concept implements a dual fluid thermodynamic cycle integrating heat exchangers into the core flow of the engine. An evaporator preheats, evaporates and superheats liquid water by cooling down the turbine exhaust flow. The generated steam is then injected into the core flow of the engine increasing efficiency and specific power of the engine. Heat Exchangers with phase change phenomena on at least one fluid side pose a novel component for aircraft engine industry. Therefore, suitable models in terms of low-fidelity tools are currently developed in order to assess the performance of such a novel component in a robust and numerically efficient way. At the DLR Institute of Propulsion Technology, a preliminary design tool for heat exchanger performance evaluation for aero engine applications is being developed (PreHEAT). A cell-based 0D P-NTU approach is applied to predict the thermodynamic performance of the heat exchanger setup. Literature correlations for both fluid sides are implemented determining the pressure loss and the transferred heat of the heat exchanger. The correlation-based prediction of the heat exchanger's thermodynamic performance is accompanied by uncertainties. In order to reduce the uncertainty of the heat exchanger's performance prediction using numerical methods, CFD results need to be fed back into the predesign tool. Therefore, in this paper, an approach is presented to couple the existing preliminary design tool with high-fidelity CFD results of the exhaust-side of the evaporator. The goal is to identify a suitable and numerically efficient method to feed back CFD results into PreHEAT and to integrate updated predesign results into the CFD simulation. The implemented coupling approach is finally demonstrated for a design point of the WET concept showing efficient and fast convergence of the coupled analysis.