An Investigation of Fan-shaped Film Cooling in High Pressure Turbine Using RANS Simulation and Neural Network Method

This is the authors' abstract. We don't add key points for this paper.

Film Cooling is used to protect the high-pressure turbine blade in advanced aero engines from high-temperature gases. The numerical simulation outcome and the accuracy of RANS simulation are significantly affected by mesh size, mesh quality, as well as flow conditions, and geometric parameters, etc. In many of these deep representations help the model current studies and design optimizations, Artificial Intelligence (AI) is deployed with deep machine learning, with dataset often generated based on CFD simulations of the cooling effectiveness of film cooling holes. A revisit of the influencing factors is thus necessary in the numerical modelling of the fan-shaped film cooling holes. In the present work, firstly, a variety of flow and geometric parameters are studied including the blowing ratio, injection angle, laidback angle, and the lateral angle, etc. The optimum blowing ratio at which best cooling efficiency is reached differs for different geometries of the hole. The various design parameters and flow conditions of fan-shaped hole are interrelated and comprehensive considerations must be carefully carried out. Secondly, the numerical results are used as input datasets for several neural network models. Different neural network architectures are used to predict the film cooling effectiveness. The outcomes demonstrate that a well-designed neural network model can be used to predict film cooling with high accuracy within a tiny fraction of computing time, even though the dataset is small at present. Finally, the sparrow search algorithm integrated with the GRU model was adopted to optimize the parameters of the cooling hole within the design space. This again indicates that neural network method is a powerful tool with great potential for direct numerical design and optimization of massive number of film cooling holes on an entire blade.