Abstract
"Propulsion airframe integration offers a range of benefits, while opening the realm of design options for future air-vehicle configurations. The aerodynamic complexity resulting from interaction between airframe and propulsion system, however, still lacks understanding. Current modeling approaches either neglect the aerodynamic interdependence by treating external vehicle aerodynamics and propulsion aerodynamics separately, or by applying non-valid assumptions on the behaviour of boundary layer during boundary layer ingestion (BLI). This work develops an aerodynamic modeling approach capable of accurately predicting the flow-field around a fully installed propulsion system, in-line of an axisymmetric body, ingesting its body boundary layer. This is accomplished by developing an empirical model of velocity component profiles at the fan inlet and exit which is then capable of delivering adaptive engine boundary conditions for CFD. The empirical model in turn is developed from experimental measurements. A low-cost modular experimental wind tunnel model is developed that allows for parametric investigation of various propulsion system shape variations. The experimentally generated flow-field data is then used for validation of CFD and the empirical model. Comparison between the experimentally generated data of the flow-field and surface pressures agrees well with CFD. A full net-thrust and drag prediction through a control volume analysis is performed and agrees well with experiment. This is evidence that accurate while efficient modeling of an air-vehicle configuration employing boundary layer ingestion is possible. Analysis of the boundary layer ingested showed that assessment of vorticity serves as a more accurate metric for definition of boundary layer edge than free stream velocity in these particular flow conditions. The presented modeling approach further highlights discrepancies between conventional modeling of boundary layer ingesting propulsion systems by comparison with the newly presented approach. Viscous body drag on the boat-tail upstream of the fan is thereby miscalculated by conventional methods by up to 8%; whereas pressure drag contributions are under-predicted by up to 28% for the cases compared. The numerical model is further coupled with a gradient based optimizer to showcase capability for shape and performance optimization. For the first time optimization for an installed BLI configuration with experimentally validated adaptive engine boundary conditions is presented."