论文部分内容阅读
以50°弯角多圆弧单列叶栅和串列叶栅为研究对象,应用数值模拟手段对所设计的两类叶栅进行数值分析,获取了其不同来流马赫数条件下的性能和流动细节。应用NSGAⅡ遗传算法结合BP神经网络技术,对串列叶栅的五个关键几何参数进行优化设计,验证了优化方法的可行性。研究结果表明:在全攻角范围内,串列叶栅的静压比都高于单列叶栅,负攻角范围内,串列叶栅损失低于单列叶栅;经过优化,串列叶栅在大负攻角下的性能略有降低,同时改善了正攻角性能,在4°攻角、0.8马赫数时静压比提升4.3%,总压损失系数降低42%;优化后串列叶栅在全工况范围内性能都要优于单列叶栅,并且串列叶栅最大压比点和最小损失点攻角均向右漂移2°。
Taking 50 ° curved multi-arc single cascade cascades and tandem cascades as the research object, numerical simulation is used to analyze the two types of cascades, and their performances and flows under different flow Mach numbers are obtained. detail. Using NSGA Ⅱ genetic algorithm and BP neural network technology, the five key geometric parameters of tandem cascade are optimized and the feasibility of the optimization method is verified. The results show that the static pressure ratio of tandem cascade cascade is higher than that of single cascade cascade in the range of full angle of attack, and the loss of tandem cascade is lower than that of single cascade in the range of negative attack angle. After optimization, At slightly negative angle of attack, the performance is slightly reduced, while the positive angle of attack performance is improved. The static pressure ratio is increased by 4.3% and the total pressure loss coefficient is reduced by 42% at the attack angle of 4 °, 0.8 Mach. After optimization, The performance of the grid is better than that of the single-cascade cascade in all operating conditions, and both the maximum pressure ratio point of the cascades and the angle of attack of the minimum loss point drift to the right by 2 °.