DTW based quantitative fault diagnosis of gas path component in turbofan
-
摘要: 提出了一种利用动态过程信息对气路故障进行定量诊断的新方法.采用动态时间规整(DTW, Dynamic Time Warping)技术,对动态测试样本同已知的故障模式进行相似性度量,充分利用发动机的过渡态特征信息,实现了发动机气路部件故障的定量诊断.结果表明:这种方法的漏报率和误报率低,具有很高的诊断准确性,对测量噪声和其它不确定性具有很好的鲁棒性.Abstract: A quantitative fault diagnosis method was presented based on pattern recognition principles for gas path component in turbofan. The method consists of a preprocessing step for multivariate dynamic data, where the magnitude dependent information is standardized, and a similarity assessment step via DTW (dynamic time warping). DTW is a flexible pattern matching method used in the area of speech recognition. The method was designed to classify faults independently of inconsistencies in the operating processes of turbofan ground test-runs. Quantitative diagnosis was fulfilled by the subdivision of fault pattern database according to the fault degree of gas path component. Case studies show that the method has a low false alarm rate and a high capability in isolating faults, and moreover, it is also robust with measuring noise and other uncertainties.
-
Key words:
- aircraft engines /
- fault diagnosis /
- dynamic process /
- dynamic time warping
-
[1] Duyar A,Eldem V,Merrill W, et al. Fault detection and diagnosis in propulsion systems:a fault parameter estimation approach[J]. AIAA Journal of Guidance, Control and Dynamics, 1994,17(1):104~108 [2] 孙 斌,张 津. BPN在涡扇发动机气路故障诊断中的应用[J]. 航空动力学报,1998,13(3):323~326 Sun Bin,Zhang Jin. Application of BPN to quantitative diagnosis of gas path component faults in turbofan[J]. Journal of Aerospace Power, 1998,13(3):323~326(in Chinese) [3] 叶志锋,孙健国. 应用神经网络诊断发动机气路故障的前景[J]. 推进技术,2002,23(1):1~4 Ye Zhifeng,Sun Jianguo. Prospect for neural networks used aeroengine fault diagnosis technology[J]. Journal of Propulsion Technology, 2002,23(1):1~4(in Chinese) [4] 周东华,叶银忠. 现代故障诊断与容错控制[M].北京:清华大学出版社,2000 Zhou Donghua,Ye Yinzhong. Modern fault diagnosis and fault-tolerant control .Beijing:Tsinghua University Press,2000(in Chinese) [5] Kassidas A,Taylor P A,MacGregor J F. Off-line diagnosis of deterministic faults in continuous dynamic multivariable processes using speech recognition methods[J]. Journal of Process Control, 1998,8(5/6):381~393 [6] Keogh E J,Pazzani M J. Scaling up dynamic time warping for data mining applications . In:Proceedings of the 6th International Conference on Knowledge Discovery and Data Mining . Boston, 2000.285~289 -
点击查看大图
计量
- 文章访问数: 2751
- HTML全文浏览量: 243
- PDF下载量: 7
- 被引次数: 0
下载:
