基于POD_Kriging代理模型的结冰云雾参数快速计算方法研究

Fast Calculation of Icing Cloud Parameters Based on POD_Kriging Surrogate Model

掌握结冰云雾参数的分布对于飞机结冰分析具有重要意义。为了获得这些参数,传统的高保真数值模拟技术计算效率低,难以应用于需要实时评估结冰的工程场景。为了克服这一困难,通过本征正交分解(Proper orthogonal decomposion,POD)方法与Kriging技术相结合,提出了一种非侵入式结冰云雾参数计算代理模型。该模型以4个飞行参数和2个云参数作为输入变量,模型输出为液体水含量(Liquid water content,LWC)和液滴收集系数。采用准三维NACA0012翼型验证了模型的准确性和计算效率。4个测试案例的数值结果表明:POD_Kriging模型准确、有效。相比于经典的FENSAP‐ICE软件,所开发的代理模型能够更有效的得到结冰云雾场。

The distribution of ice cloud parameters is of great significance to the analysis of aircraft icing.In order to obtain these parameters,traditional high-fidelity numerical simulation techniques have low computational efficiency and are difficult to apply in engineering scenarios that require real-time evaluation of icing.To overcome this difficulty,a non-intrusive surrogate model is proposed by proper orthogonal decomposion(POD)model coupled with the Kriging technology.The model takes four flight parameters and two cloud parameters as input variables.The model outputs are liquid water content(LWC)and droplet collection efficiency.A quasi-three-dimensional NACA0012 airfoil is adopted to verify the accuracy and computational efficiency of the model.The numerical results of the four test cases show that the POD_Kriging model is able to produce satisfactory results for the statistical quantities of interest.It is found that the developed surrogate model is computationally more efficient than the classical FENSAP-ICE for simulation of icing cloud field.