This paper describes the combination of a Kriging surrogate model with a micro-genetic algorithm for studying the influence of circumferential perturbations from the stack line on the high-cycle fatigue (HCF) weak link of a fan blade in a high-bypass-ratio turbofan engine. Based on the circumferential perturbations of the fan blade stack line, an automated system is developed for the parameterized modeling and meshing of the blade, and an integral platform is established for parameterized modeling, finite-element simulation, and optimization. The static stress, index of strain energy density, and vibratory stress margin of the 1st flex mode are set as the objective functions for the optimization of a fan blade. The results show that the optimized blade has three areas of low static stress, unlike the single “bull’s eye” distribution of the baseline blade. Optimization reduces the maximum static stress by 5.87% and the strain energy density index by 0.77%, while increasing the vibratory stress margin under the same dynamic load by 9.51%. The natural frequencies, mode shapes, resonance margins, and aerodynamic parameters exhibit no significant changes, which illustrates that the optimization method can improve the static stress and vibratory stress distribution of the fan blade without negatively influencing the other vibration and aerodynamic characteristics. The proposed method is an effective means of fan blade design optimization, and could be applied alongside other design variables and objective functions, such as the swept and skewed configuration of the stack line and twist angle of the blade, to optimize the vibration characteristics and aero-elastic performance.

本文介绍了Kriging替代模型与微遗传算法的组合，用于研究烟囱流水线的圆周扰动对高旁路比涡轮风扇中风扇叶片的高周疲劳（HCF）薄弱环节的影响发动机。基于风扇叶片堆叠线的圆周扰动，开发了用于叶片参数化建模和啮合的自动化系统，并建立了用于参数化建模，有限元模拟和优化的集成平台。将第一挠曲模式的静态应力，应变能密度指数和振动应力裕度设置为优化风扇叶片的目标函数。结果表明，优化后的叶片具有三个低静应力区域，不同于基准刀片的单一“靶心”分布。优化使最大静态应力降低了5.87％，应变能密度指数降低了0.77％，而在相同动载荷下的振动应力裕量提高了9.51％。固有频率，振型，共振容限和空气动力学参数均无显着变化，这表明该优化方法可以改善风扇叶片的静应力和振动应力分布，而不会负面影响其他振动和空气动力学特性。所提出的方法是优化风扇叶片设计的有效方法，可以与其他设计变量和目标函数（例如烟囱线的扫掠和偏斜配置以及叶片的扭曲角度）一起应用。