首页|Nonlinear uncertainty impact of geometric variations on aerodynamic performance of low-pressure turbine blades with ultra-high loading under extreme operational conditions
Nonlinear uncertainty impact of geometric variations on aerodynamic performance of low-pressure turbine blades with ultra-high loading under extreme operational conditions
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
万方数据
维普
Uncertainty impact of random geometric variations on the aerodynamic performance of low-pressure turbine blades is considerable,which is further amplified by the current ultra-high-lift design trend for weight reduction.Therefore,this uncertainty impact on ultra-highly loaded blades under extreme operational conditions near the margins with potential large-scale open separation is focused on in this study.It is demonstrated that this impact is significant,unfavourable,and non-linear,which is clearly severer under extreme conditions.In addition to the overall attenuation and notable scattering of specific performance,the operational margins with open separation are also notably scattered with great risk of significant reduction.This scattering and nonlinearity are dom-inated by the variations in leading-edge thickness.The thinning of leading edge triggers local tran-sition,enhancing downstream friction and reducing resistance to open separation,which is further exacerbated by operational deterioration.However,the opposite thickening yields less benefit,implying nonlinearity.This unfavourable impact highlights the need for robust aerodynamic design,where both a safer operational condition and a more robust blade are indispensable,i.e.,a compro-mise among performance,weight,and robustness.Besides the necessary limitation of loading levels,a mid-loaded design is recommended to reduce adverse pressure gradients in both the leading edge and rear region of the suction side,which helps to decrease the susceptibility of the transition and open separation to random perturbations.Similar improvements can also be achieved by appropri-ately thickening the leading edge.
万方数据
维普
