The invention discloses an airfoil separation flow prediction method based on flow field inversion and machine learning, and the method comprises the following steps: carrying out the flow field inversion of the separation flow of an S809 airfoil at a large angle of attack, and obtaining a correction coefficient for an SA turbulence model; an SA turbulence model is extracted to solve convergent flow field features, a machine learning method is utilized to establish a graph neural network framework, and a mapping relation between the flow field features and turbulence model correction coefficients is established; and predicting the correction coefficient of the turbulence model by using the trained graph neural network model. According to the method, a small amount of experimental data is utilized, correction aiming at the turbulence model is obtained through flow field inversion, the model is trained in combination with a machine learning method, the model can predict correction of the RANS turbulence model under other flow conditions, the step of flow field inversion is omitted, and the accuracy of flow field inversion is improved. The efficiency is improved; and meanwhile, the effect of improving the inaccuracy of the turbulence model on separation flow prediction is achieved.
本发明公开了一种基于流场反演和机器学习的翼型分离流动预测方法,包括以下步骤:对S809翼型大攻角下的分离流动进行流场反演,得到针对SA湍流模型的修正系数;提取SA湍流模型求解收敛的流场特征,并利用机器学习方法搭建图神经网络框架,建立流场特征与湍流模型修正系数间的映射关系;利用训练好的图神经网络模型对湍流模型的修正系数进行预测。本发明利用少量的实验数据,通过流场反演得到针对湍流模型的修正,再结合机器学习方法训练模型,使得模型能够预测其它流动情况下RANS湍流模型的修正,省略了流场反演的步骤,提升效率的同时达到改进湍流模型对于分离流动预测不准确的效果。
Airfoil profile separation flow prediction method based on flow field inversion and machine learning
一种基于流场反演和机器学习的翼型分离流动预测方法
2023-12-22
Patent
Electronic Resource
Chinese
IPC: | G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / B64F GROUND OR AIRCRAFT-CARRIER-DECK INSTALLATIONS SPECIALLY ADAPTED FOR USE IN CONNECTION WITH AIRCRAFT , Boden- oder Flugzeugträgerdeckeinrichtungen besonders ausgebildet für die Verwendung in Verbindung mit Luftfahrzeugen / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen |
Airfoil Profile in a Nonuniform Flow
NTIS | 1978
|Airfoil profile in a nonuniform flow
NTRS | 1978
|Dynamic Airfoil Flow Separation and Reattachment
Online Contents | 1995
|European Patent Office | 2023
|Dynamic airfoil flow separation and reattachment
AIAA | 1995
|