The secondary flow in compressor passages contribute significantly to the decline of efficiency. To im prove the compressor efficiency, the secondary flow can be controlled by changing the hub end wall structure. A 3-D turbomachinery optimization design method based on artificial neural network and genetic algorithm was applied. The optimization objective was keeping the mass flow unchanged, the pressure rate was no less than before at the peak ef ficiency operating point. The optimization result presented a new non-axisymmetric end wall which reduced the total pressure loss in the rotor passage, suppressed the corner separation in the downstream stator passage, and made the compressor efficiency improved. The results showed that the non-axisymmetric end wall can effectively control flow loss near end wall and improve compressor efficiency.%关于优化提高压气机效能问题,由于压气机通道的二次流流动会造成流动损失,引起效率下降.为了解决上述问题,通过改变轮毂端壁结构,可以控制二次流流动,以提高压气机效率.采用人工神经网络及遗传算法的叶轮机械三维优化建模方法,使在最高效率工况下可以保持流量不变,压比不低于优化前.对压气机转子轮毂结构进行了优化,得到新型非轴对称端壁结构,并进行仿真.结果表明,降低了转子通道内的相对总压损失,抑制了下游静叶角区分离,可使压气机提高效率,并能有效控制端壁附近的流动损失,提高压气机效率.
展开▼
