叶栅气动反问题的伴随优化解法及应用

Adjoint Optimization Method for Aerodynamic Inverse Problem of Cascades

伴随方法是目前流体机械优化设计领域的研究热点,具有计算量与设计变量数目基本无关的优点。鉴于以往相关研究尚未将伴随方法用于有分离流动条件下的叶栅气动反问题设计,建立了一套集叶片几何参数化、网格生成、流场求解、伴随场求解与优化求解于一体的叶栅气动反问题的优化求解方法,从减弱流动分离的角度出发,通过给定更合适的叶片表面压力分布,完成了叶栅反问题求解。研究表明,所得叶片吸力面型线更为平缓,在所研究的2种攻角下的尾缘附近流动分离较优化前得到了有效缓解。该研究有利于发展高效、宽工况叶栅设计技术,并可为复杂流体机械部件的先进设计奠定理论基础。

Adjoint method based design optimization is currently a hot topic in the field of fluid machinery, which has the advantage that the computation cost is almost independent of the number of design variables. In view of that few efforts in previous researches have been devoted to the inverse aerodynamic design of cascades with flow separation by means of the adjoint method, an optimization method for aerodynamic inverse problem of cascade, which includes blade geometry parameterization, grid generation, flow solving, adjoint field solving and optimizing, is proposed. From the point of view of weakening the flow separation, the aerodynamic inverse design of a cascade is completed by giving a desirable surface pressure distribution on the blade. The results show that the suction profile of the obtained blade becomes flatter than the initial design, and the flow separation near the trailing edge is weakened at two investigated attack angles. This approach is beneficial for the development of aerodynamic design technology of cascades with higher efficiency and wider operating range.