Low-pressure turbine blade profile aerodynamic design method based on optimal load distribution model optimization

Abstract

The invention discloses a low-pressure turbine blade profile aerodynamic design method based on optimization of an optimal load distribution model, and the method comprises the steps: generating a blade profile geometric sample according to a given blade profile parameter value range and low-dimensional aerodynamic design parameters, calculating the optimal load distribution through CFD when the total pressure loss of the blade profile is minimum, and generating a blade profile database; constructing an optimal load distribution model based on a multi-output Gaussian process and a deep neural network, and training the optimal load distribution model by minimizing a marginal likelihood loss function according to the training samples in the blade profile database to obtain a hyper-parameter group in the optimal load distribution model; and according to the trained optimal load distribution model, target optimal load distribution of the target low-dimensional aerodynamic design parameters is calculated, and an optimal aerodynamic blade profile corresponding to the target optimal load distribution is calculated by using a blade profile inverse design model. According to the design method, the precision and efficiency of pneumatic design of the turbine blade profile can be improved, and the geometric design period of the turbine blade profile can be shortened.

Description

technical field [0001] The invention relates to the technical field of turbine blade design, in particular to an aerodynamic design method for a low-pressure turbine blade based on optimal load distribution model optimization. Background technique [0002] The turbine is a key component of an aero-engine. It is a turbomechanical device that converts the energy of high-temperature and high-pressure gas into kinetic energy and mechanical energy. The aerodynamic performance of its components has a very important impact on the overall aerodynamic performance, economy and environmental protection of the engine. [0003] Turbine aerodynamic design usually faces complex design requirements, and aerodynamic performance is very sensitive to geometric changes. Many design requirements and geometric variables are coupled together to form a high-dimensional design space, making it difficult to obtain the corresponding global optimum from theoretical quantitative analysis. Therefore, the...

AI Technical Summary

Problems solved by technology

[0003] Turbine aerodynamic design usually faces complex design requirements, and aerodynamic performance is very sensitive to geometric changes. Many design requirements and geometric variables are coupled together to form a high-dimensional design space, making it difficult to obtain the corresponding global optimum from theoretical quantitative analysis. However, it can only simulate and evaluate the aerodynamic performance of turbine blades based on experiments or computational fluid dynamics (Computational Fluid Dynamics, CFD) values, and rely on experts or use gradient or stochastic optimization methods to iteratively optimize the geometric parameters of the blade.

[0004] However, the existing aerodynamic design methods usually require high-precision CFD iterative calculations. For example, in the patent application with the publication number CN112380794A, a multi-disciplinary parallel collaborative optimization design method for aero turbine engine blades is disclosed. This type of design method is computationally expensive. , The design cycle is long, and most of them are optimized based on the gradient method, which is easy to fall into the local minimum

[0005] Although the surrogate model based on neural network or Gaussian process regression can be used to reduce the time consumption of CFD calculation, the calculation accuracy of such surrogate model is relatively poor, and with the gradual increase of design variables in the surrogate model, such The surrogate model algorithm faces a serious dimensionality disaster problem, which makes it impossible to establish a general surrogate model between a wide range of design conditions and airfoil geometry and aerodynamic performance

Images