航空学报 > 2024, Vol. 45 Issue (20): 230098-230098   doi: 10.7527/S1000-6893.2024.30098

基于宽频振动的燃气轮机叶片断裂故障识别特征与辨识方法

胡明辉1,2, 刘少朋3, 王浩4, 李晨阳2, 王维民1,3, 江志农1,2()   

  1. 1.北京化工大学 高端压缩机及系统技术全国重点实验室,北京 100029
    2.北京化工大学 发动机健康监控及网络化教育部重点实验室,北京 100029
    3.北京化工大学 高端机械装备健康监控与自愈化北京市重点实验室,北京 100029
    4.中国舰船研究院,北京 100101
  • 收稿日期:2024-01-03 修回日期:2024-01-30 接受日期:2024-03-18 出版日期:2024-10-25 发布日期:2024-03-19
  • 通讯作者: 江志农 E-mail:2003500007@buct.edu.cn
  • 基金资助:
    国家自然科学基金重点项目(92160203);特殊领域青年人才托举工程项目(2022-JCJQ-QT-059);博士后创新人才支持计划(BX20180031)

Characterization and identification of gas turbine blade fracture faults based on broadband vibration

Minghui HU1,2, Shaopeng LIU3, Hao WANG4, Chenyang LI2, Weimin WANG1,3, Zhinong JIANG1,2()   

  1. 1.State Key Laboratory of High-end Compressor and System Technology,Beijing University of Chemical Technology,Beijing  100029,China
    2.Key Laborotory of Engine Health Monitoring-Control and Networking of Ministry of Education,Beijing University of Chemical Technology,Beijing  100029,China
    3.Beijing Key Laboratory of Health Monitoringlf and Self-Recovery for High-End Mechanical Equipment,Beijing University of Chemical Technology,Beijing  100029,China
    4.China Ship Research and Development Academy,Beijing  100101,China
  • Received:2024-01-03 Revised:2024-01-30 Accepted:2024-03-18 Online:2024-10-25 Published:2024-03-19
  • Contact: Zhinong JIANG E-mail:2003500007@buct.edu.cn
  • Supported by:
    the Key Program of National Natural Science Foundation of China(92160203);Young Elite Scientists Sponsorship Program by CAST(2022-JCJQ-QT-059);National Postdoctoral Program for Innovation Talents Support Program(BX20180031)

摘要:

针对燃气轮机动叶片断裂故障难以及时、有效告警与识别的技术难关开展了研究。从叶片断裂故障机理出发,分析了不平衡转子对尾流压力的调制作用,建立了叶片激振力模型和激振力作用下机匣的强迫振动响应模型,求解了动叶片断裂激励下的宽频振动响应规律。在此基础上,构建了叶片断裂故障敏感特征并提出了基于改进K近邻(O-KNN)的故障告警模型,实现叶片断裂故障的及时告警分析。进一步提出了叶片断裂故障定位参数,通过各级叶片特征参数的对比实现断裂故障的精准定位。综合以上研究,构建了宽频振动特征驱动的燃气轮机叶片断裂故障辨识方法,并通过某型燃气轮机实测案例验证了故障敏感特征及辨识方法的有效性。

关键词: 燃气轮机, 叶片断裂, 故障诊断, 机匣振动, O-KNN

Abstract:

Aiming at the technical difficulty of timely and effective warning and identification of gas turbine rotor blade fracture faults, this paper conducts a three-pronged research. Starting from the blade fracture fault mechanism, we analyse the modulation effect of the unbalanced rotor on the wake pressure, establish the blade excitation force model and the forced vibration response model of the casing under the action of the excitation force, and solve the wide-frequency vibration response of the dynamic blade fracture excitation. On this basis, a blade fracture fault sensitive characteristic is constructed, and a fault alarm model based on the Optimized K-Nearest Neighbor (O-KNN) is proposed to realize the timely alarm analysis of the blade fracture fault. Furthermore, the blade fracture fault localization parameters are proposed, and the accurate localization of the fracture fault is realized by comparing the blade characteristic parameters at all stages. In summary, a broadband vibration characteristic driven gas turbine blade fracture fault identification method is constructed, and a test case on a certain type of gas turbine verifies the effectiveness of the fault sensitive characteristics and the identification method.

Key words: gas turbine, blade fracture, fault diagnosis, casing vibration, optimized K-nearest neighbor

中图分类号: