Gas-Path Optimization Using Turbine Aerodynamics Meanline and Design Exploration

One of the steps in the complex design of aircraft propulsion system consists of designing an efficient turbine gas - path . Gas - path geometry affects how the stators, rotors and duct will be designed. For this reason, the optimization of the gas - path needs to be done early in the development process of an engine. This is the concept of Preliminary Multi - Disciplinary Optimization (PMDO). At this stage of the design process, as many configurations as possible should be analysed. Due to the size of the design space to be explore d thousands of configurations have to be considered. Consequently, each of these configurations needs to be simulated rapidly. To meet this requirement , an in - house 1D meanline code based on a correlation loss model is used. The optimization is done with a combination of direct optimization and design exploration. A simple direct optimization is used to generate a first version of the gas - path. Design exploration is achieved with an in - house Framework for Design Exploration (FDE). This framework includes Design of Experiment (DOE) and Surrogate Assisted Optimization (SAO) workflow. SAO is executed to find a global optimum configuration within user defined limitations. These design limits are de fined according to a set of predefined limitations on factors (in put parameters) and constraints on responses (output parameters). Furthermore, any responses can be set as an objective to be minimized, maximized or to be targeted for a specific value. To obtain such an optimized gas - path, a robust parametrization has to be developed. An efficient parametrization will limit the number of nonphysical gas - path configurations in the design space without excluding any optimal configurations. This work focuses on optimizing the turbine gas - path to achieve one of three possible objectives. For most of the cases, the objective is maximizing efficiency. Another possible objective consists of minimizing the total length of the turbine while achieving a specific efficiency. Finally, this optimization tool can be used by turbine aerodynamicists to quickly analyse different stage configurations such as two versus three Power Turbines (PT). Iterating on the number of PT stages becomes much faster when the entire optimization process is automated