基于相似匹配的液体火箭发动机故障模式挖掘
详细信息 本馆镜像全文    |  推荐本文 | | 获取馆网全文
摘要
将时间序列相似性匹配方法引入到液体火箭发动机故障模式挖掘中。针对发动机试车数据的特点,提出了一种基于序变换的时间序列相似匹配算法。该算法具有对时间序列幅值和持续时间不敏感、抗噪声能力强等优点。对某型液体火箭发动机故障数据的相似匹配实验表明:该算法能够为液体火箭发动机的故障检测和诊断提供较好的技术支持。
The method of time series similarity matching (TSSM) is introduced in the fault mode mining of the liquid rocket engine (LRE). According to the feature of fire-test data, an ordinal transformation based algorithm is proposed, which is insensitive to amplitude, time span and noises of the time series. The TSSM experiments on the data of some LRE indicated that it is quite supportive to fault detection and diagnosis.
引文
[1]张惠军.液体火箭发动机故障检测与诊断技术综述[J].火箭推进,2004,30(5):40-45.
    [2]张振鹏.液体火箭发动机故障检测与诊断中的基础问题研究[J].推进技术,2002,23(5):353-359.
    [3]陈启智.液体火箭推进系统健康监控技术的演变[J].推进技术,1997,18(1):1-7.
    [4]胡小平,张丽娟,王艳梅,等.液体火箭发动机故障检测和诊断中数据挖掘策略的分析[J].国防科技大学学报,2005,27(3):1-5.
    [5]Agrawal C,Faloutsos A.An efficient Similarity Search in Sequence Database[C].Proceedings ofthe4th International Conference of Foundations of Data Organization and Algo-rithms,Chicago,October1993:69-84.
    [6]吴绍春,吴耿锋,王炜,等.寻找地震相关地区的时间序列相似性匹配算法[J].软件学报,2006,17(2):185-192.
    [7]Li Qiu-Dan,Chi Zhong-Xian,Wang Zhan-Chang.A Novel Method for Similarity Search Over Electric Time Series Data[C].International Conference on Machine Learning and Cybernetics,Beijing,China,2-5Nov.2003(4):2315-2318.
    [8]Fu Wai-chee,Eamonn Keogh,Leo Yung Hang.Scaling and Time Warping in Time Series Querying[C].Proceed-ings of the31st VLDB Conference,Trondheim,Norway,2005:649-660.
    [9]Keller K,Sinn M.Ordinal analysis of time series[J].Phys-ica A:Statistical Mechanics and its Applications(Ger-many)Vol.356,No.1,2005:114-120.
    [10]张育林,吴建军,朱恒伟,等.液体火箭发动机健康监控技术[M].长沙:国防科技大学出版社,1998.

版权所有:© 2023 中国地质图书馆 中国地质调查局地学文献中心