Coordinated application of CFD and gas turbine performance methods

Abstract

In conventional gas turbine performance methods, components are represented by characteristics where the 3d properties of the flow fields' are averaged providing key flow properties at component interfaces. For the design of a component, the consideration of the 3d nature of the flow is crucial and Computational Fluid Dynamics (CFD) is well established in the design process; however computational challenges generally limit the application to individual components only. The fact that each engine component has an effect on the overall engine performance sets a strong incentive to link the two methods closer together. This thesis explores the potential benefits of the 'coordinated application of performance modelling and CFD through two proof-of-concept case studies. The, research context of the first case study was high fidelity performance simulation. A performance simulation process With a 2d radial representation of the low pressure system has been developed for a high bypass turbofan. The intake and bypass section were represented using axial-symmetric 2d CFD modules and radial flow profiles were exchanged at the component interfaces to an existing 2d radial fan model. The iteration procedure between the two tools was performed manually and investigated for various operating conditions. A match of the flow data between the tools was achieved for the intake. Guidance for improvements and an automation of the process are also given. The research of the second case study involved the design of component subsystems in an industrial application and illustrates the potential of using CFD and performance simulation for the design of a compressor washing system; The optimisation of a compressor washing system for an industrial gas turbine has been studied using a 3d CFD model of the intake component. Due to the interaction of droplets with the surrounding air flow, the key factors from injection system and' the air flow demand of the gas turbine were investigated. Boundary conditions were provided using a performance simulation tool which facilitated the evaluation of relevant operating conditions. Associated spray parameters were derived from numerical sensitivity studies calibrated to data from visual field inspections and key design parameters were provided. Improvements were suggested and confirmed for a similar field installation. ~