基于数值模拟的流场附面层边缘识别方法

附面层边缘通常取在速度达到主流速度0.99倍的位置,而复杂流场中主流流动往往并不均匀,给附面层边缘的准确识别造成了困难。为解决此问题,提出了用"参考主流"代替实际主流识别附面层边缘的方法:通过零剪切力滑移壁面边界条件下数值模拟得到不受附面层干扰的参考主流,在根据附面层定义确定附面层边缘时以该参考主流中的速度代替实际的主流速度。通过斜楔压缩和弯曲压缩两个超声速压缩流场对该识别方法进行了验证,所得到的斜楔压缩出口截面上附面层厚度与采用实际主流速度判断得到的厚度相对误差仅4.1%。根据该方法的识别结果对弯曲压缩型面设计进行附面层修正后,弯曲激波高度与无黏设计值之间的误差从修正前的2.0%降低至0.3%,压缩面末端压力的相对误差从修正前的6.6%降低至2.3%。该方法避免了指定主流速度的主观性,识别结果较为准确。

Flow boundary layer edge detection method based on numerical simulation

Boundary layer edge is usually defined as the position where the velocity is 0.99 times of the main flow.However,the main flow in complicated flow field is often non-uniform,which makes it difficult to accurately identify the boundary layer.To solve this problem,a boundary layer edge detection method is proposed by using"reference mainstream"instead of the actual mainstream.The reference main flow which is not disturbed by the boundary layer is obtained by numerical simulation under zero shear stress slip wall boundary condition,and the actual main flow velocity replaced by velocity at corresponding position in the reference main flow when the boundary layer edge is determined according to the boundary layer definition.Two supersonic compression flow fields including multi-ramp compression and curved surface compression are used to evaluate this detection method.The results show that relative error between the thickness of boundary layer at multi-ramp compression outlet station and the actual mainstream velocity is only 4.1%.According to the identification results of the method,the error between the bending shock height and the inviscid design value is reduced from 2.0%to 0.3%,and the relative error of the end pressure of the compression surface is reduced from 6.6%to 2.3%.This detection method could avoid the arbitrariness when determining the main flow velocity,and the recognition result is accurate.