Method for predicting Reynolds anisotropy state distribution of flow around near-wall cylinder

The invention discloses a near-wall cylinder streaming Reynolds anisotropy state distribution prediction method. The method comprises the following steps: step 1, establishing a near-wall cylinder streaming CFD geometric model; 2, extracting input features; step 3, extracting output features; and 4, constructing a prediction model by using a random forest algorithm. According to the method, the calculation precision of the Reynolds stress model can be achieved by only needing calculation conditions required by the vortex viscosity model. 本发明公开了一种近壁圆柱绕流雷诺各向异性状态分布的预测方法，包括如下步骤：步骤1：建立近壁圆柱绕流CFD几何模型；步骤2：提取输入特征；步骤3：输出特征提取；步骤4：利用随机森林算法构建预测模型。本发明方法只需涡粘模型所需计算条件便可达到雷诺应力模型的计算精度。