分形维数下航空发动机风扇叶型侵蚀前缘形貌几何的参数化  

Study on the Parameterization of the Erosion Leading Edge Morphology of Aircraft Engine Fan Blades in Fractal Dimension

作　　者：刘文然 史磊 Liu Wenran;Shi Lei(Sino-Europe Institute of Aviation Engineering,Civil Aviation University of China,Tianjin 300300)

机构地区：[1]中国民航大学中欧航空工程师学院,天津300300

出　　处：《中阿科技论坛(中英文)》2025年第2期108-112,共5页China-Arab States Science and Technology Forum

基　　金：2024年中央高校基本科研业务费中国民航大学专项“前缘侵蚀风扇转子叶型气动特性衰变机理分析及衍化过程研究”(3122024028)。

摘　　要：航空发动机的风扇转子叶片在经历长时间运作后,其前缘形貌会因遭受风蚀及砂石颗粒的冲击侵蚀而发生显著变化。为了刻画航空发动机风扇叶片侵蚀前缘形貌的特征,文章利用分形几何学知识,克服测量仪器在精度上的局限性,基于精密测量数据计算出叶片前缘形貌的分形维数,以此为基准对数据点进行简化处理,再通过构建并训练BP神经网络模型,实现数据点稠密化,从而精细且准确地描述叶片前缘形貌,最后采用分段拟合与粘合的方法,完成叶片前缘形貌几何的参数化。实践证明,该参数化方法不仅有助于提升形貌刻画的精度,也能为航空发动机叶片的维护、修复及性能优化提供有力的技术支持。After prolonged operation,the leading edge morphology of fan rotor blades in aircraft engines will undergo significant changes due to wind erosion and impact erosion by sand and gravels.In order to characterize the erosion morphology,this article applies fractal geometry to overcome the limitations of measurement instruments in accuracy.Based on precise data,the fractal dimension is calculated,and the data points are simplified.Then,by constructing and training a BP neural network model,the data points are densified to accurately and finely describe the morphology.Finally,the geometric parameterization of the morphology is completed using segmented fitting and bonding methods.Through experiment,it has been proven that this method not only helps to enhance the accuracy of morphology characterization,but also provides technical support for the maintenance,repair and performance optimization of aircraft engine blades.

关 键 词：风扇转子 侵蚀前缘 分形维数 BP神经网络 航空发动机 

分 类 号：V241.0[航空宇航科学与技术—飞行器设计]