航空学报 > 2024, Vol. 45 Issue (21): 629385-629385   doi: 10.7527/S1000-6893.2023.29385

面向运维的HPT叶片寿命数字孪生建模方法

李春华, 孙见忠(), 陆纪龙   

  1. 南京航空航天大学 民航学院,南京 211106
  • 收稿日期:2023-07-29 修回日期:2023-08-31 接受日期:2023-10-09 出版日期:2024-11-15 发布日期:2023-11-07
  • 通讯作者: 孙见忠 E-mail:sunjianzhong@nuaa.edu.cn
  • 基金资助:
    国家自然科学基金(52072176)

Maintenance-oriented approach for HPT blade life digital twin modeling

Chunhua LI, Jianzhong SUN(), Jilong LU   

  1. College of Civil Aviation,Nanjing University of Aeronautics and Astronautics,Nanjing 211106
  • Received:2023-07-29 Revised:2023-08-31 Accepted:2023-10-09 Online:2024-11-15 Published:2023-11-07
  • Contact: Jianzhong SUN E-mail:sunjianzhong@nuaa.edu.cn
  • Supported by:
    National Natural Science Foundation of China(52072176)

摘要:

高压涡轮(HPT)叶片是典型的高温、高负荷、结构复杂的热端部件,其寿命不仅取决于设计、制造和工艺水平,还与发动机的实际运行环境、使用条件以及维修情况等密切相关,如何融合服役阶段多模态运维数据以提高HPT叶片寿命预测精度、降低寿命预测不确定性则成为一个重要问题。在基于使用的寿命管理(UBL)方法基础上,结合数字孪生技术,提出了数据和模型融合驱动的服役HPT叶片寿命数字孪生(LDT)建模方法,基于多模态运维数据表征和跟踪HPT叶片在实际服役条件下的寿命消耗及特定使用条件下的服役寿命预测。另外,还提出了LDT涉及到的不确定性量化与管理的方法。所提方法在某HPT叶片上得到了验证,结果表明提出的LDT及不确定性量化与管理方法可以有效融合多模态运维数据,降低叶片服役载荷及寿命预测的不确定性,有效提高HPT叶片寿命数字孪生模型的保真度,能够为基于数字孪生的发动机单机寿命管理及智能运维提供基础方法支撑。

关键词: 寿命数字孪生, 不确定性量化与管理, 服役HPT叶片, 运行与维护, 信息融合

Abstract:

High-Pressure Turbine (HPT) blade is a typical hot-end component with high temperature, high load and complex structure. Its in-service life depends not only on the level of design, manufacture and process, but also on the actual operating environment, operating conditions and maintenance of the engine. Therefore, how to fuse multi-modal operation and maintenance data to improve the prediction accuracy of HPT blade in-service life and reduce the prediction uncertainty is an important issue. Based on the Usage-Based Life (UBL) method and digital twin technology, this paper proposes a Life Digital Twin (LDT) modeling method for in-service HPT blades driven by data and model fusion. Based on multi-modal operation and maintenance data, the life consumption of HPT blades under actual service conditions can be characterized and tracked, and the in-service lifetime can be predicted under specific operating condition. In addition, the approach of uncertainty quantification and management involved in LDT is also proposed. The approach proposed has been verified on an HPT blade. The results show that the LDT and uncertainty quantification and management approach proposed can effectively fuse multi-modal operation and maintenance data, reduce the prediction uncertainty of HPT blade’s in-service load and life, as well as improve the fidelity of HPT blade life digital twin model; therefore, provides a basic approach support for single engine life management and intelligent operation and maintenance based on digital twin.

Key words: life digital twin, uncertainty quantification and management, in-service HPT blade, operation and maintenance, information fusion

中图分类号: