Computational fluid dynamics-based approach for low-order models of propelling nozzle performance

At the preliminary design stage for an aero-engine, the evaluation of the nozzle performance is an important aspect as it affects the overall engine cycle behaviour. Currently, there is a lack of systematic, extensive data on the nozzle performance and its dependence on the geometric and aerodynamic aspects. This paper presents a method that can be used to build characteristic maps for a nozzle as a function of a number of geometric and aerodynamic parameters. The proposed method encompasses the design of a nozzle configuration, a parameterisation of the nozzle pressure ratio, nozzle contraction ratio, plug half-angle (β), mesh generation, and an aerodynamic assessment using the Favre-averaged Navier–Stokes method. The method has been validated against experimental performance data of a plug nozzle configuration and then used for the aerodynamic assessment. The derived nozzle maps show that the thrust coefficient ( C fg ) for this type of nozzle is significantly sensitive to the combined effect of the variation of the proposed parameters on the nozzle performance. These maps were used to build low-order models to predict C fg , using response surface methods. The performance was assessed, and the results show that these low-order methods are capable of providing C fg estimates with sufficient accuracy for use in preliminary design assessments.