基于慢特征密度聚类的气路异常检测方法

孙昊; 付旭云; 钟诗胜

doi:10.13224/j.cnki.jasp.20200382

基于慢特征密度聚类的气路异常检测方法

doi: 10.13224/j.cnki.jasp.20200382

哈尔滨工业大学（威海）海洋工程学院，山东威海 264209

基金项目: 山东省自然科学基金（ZR2019MEE096）；中央高校基本科研专项资金（HIT.NSRIF.201704）；国家自然科学基金民航联合基金（U1733201）

详细信息

作者简介:
孙昊（1996-），男，硕士生，主要从事民航发动机气路异常检测。

通讯作者:
付旭云（1981-），男，副教授，博士，主要从事智能运维，预测与健康管理研究。E-mail：fuxuyun@hit.edu.cn

中图分类号: V263.4
计量
- 文章访问数: 304
- HTML浏览量: 92
- PDF量: 89
- 被引次数: 0
出版历程
- 收稿日期: 2020-09-11
- 刊出日期: 2021-10-28

Slow feature density clustering-based gas path anomaly detection method

School of Ocean Engineering，Harbin Institute of Technology，Weihai，Weihai Shandong 264209，China

摘要

摘要: 在结合核技巧、慢特征分析算法与密度聚类方法的基础上，提出了基于混合核慢特征分析和密度聚类的慢特征密度聚类算法，实现了基于民航发动机气路参数原始值的异常检测。核技巧的引入克服了慢特征分析法处理复杂数据时可能存在的维度爆炸问题，充分利用不同核函数的特点和慢特征分析的优势从气路参数原始值中提取出随时间变化最缓慢的特征作为密度聚类算法的输入，最终筛选出异常值。经实验对比发现，该方法针对某些异常拥有最好的聚类效果和最低的虚警率，尤其是检测可调放气活门系统异常时虚警数量不到样本总数量的0.5%，是一种有效的方法。
- 民航发动机 /
- 气路 /
- 异常检测 /
- 慢特征分析 /
- 核技巧 /
- 密度聚类
Abstract: Based on the kernel method，slow feature analysis algorithm and density clustering method，a mixed-kernel slow feature density clustering algorithm was proposed to detect the original gas path parameters of civil aero-engine.The introduction of the kernel method overcame the possible dimensional explosion when the slow feature analysis dealt with complex data，and it took full use of the characteristics of different kernel functions and advantages of slow feature analysis.This algorism can extract the feature with the slowest time-dependent change from the original gas path parameters and use it as the input of the density clustering algorithm.Then the anomalies were found out.Through experimental validation，this method had the best clustering results and the lowest false alarm rate on certain anomalies.Especially，the number of false alarms was less than 0.5% of the total number of samples when detecting the anomaly of the variable bleed valve system，proving it is an efficient method.
- civil aero-engine /
- gas path /
- anomaly detection /
- slow feature analysis /
- kernel method /
- density clustering

HTML

参考文献(30)

[1]	钟诗胜,李旭,张永健.基于DBN的不均衡样本驱动民航发动机故障诊断[J].航空动力学报,2019，34（3）:708-716.
[2]	QU J F,JIANG S Q,YUAN S H,et al.Analysis of an aero-engine gas path fault monitoring sensor model[J].Advanced Materials Research,2014,1016:516-520.
[3]	付松.基于深度特征挖掘的民航发动机气路故障诊断技术研究[D].哈尔滨：哈尔滨工业大学,2018.
[4]	杨洪富,贾晓亮,任寿伟.基于数据驱动的航空发动机故障诊断与预测方法综述[J].航空精密制造技术,2016,52（5）:6-9.
[5]	王奕首,余映红,卿新林,等.基于KPCA和DBN的航空发动机排气温度基线模型[J].航空发动机,2020,46（1）:54-60.
[6]	瞿红春,林文斌,许旺山,等.基于混沌PSO_Elman网络的航空发动机基线挖掘[J].中国民航大学学报,2020,38（1）:19-23,53.
[7]	林文斌.基于数据和知识驱动的发动机监控技术的研究[D].天津：中国民航大学,2019.
[8]	崔智全.民航发动机气路参数偏差值挖掘方法及其应用研究[D].哈尔滨:哈尔滨工业大学,2013.
[9]	ZHOU Xingjie,FU Xuyun,ZHAO Minghang,et al.Regression model for civil aero-engine gas path parameter deviation based on deep domain-adaptation with Res-BP neural network[J].Chinese Journal of Aeronautics,2021,34（1）:79-90.
[10]	刘翠芳,冯学晓.小型多旋翼无人机发动机气路故障残差参数偏差值挖掘[J].科技通报,2016,32（10）:184-187.
[11]	LUO H,ZHONG S.Gas turbine engine gas path anomaly detection using deep learning with Gaussian distribution[C]∥Proceedings of the 2017 Prognostics and System Health Management Conference （PHM-Harbin）.New York:Institute of Electrical and Electronics Engineers,2017:1-6.
[12]	FU Xuyun,CHEN Haibo,ZHANG Guangyao,et al.A new point anomaly detection method about aero engine based on deep learning[C]∥2018 International Conference on Sensing,Diagnostics,Prognostics,and Control.New York:Institute of Electrical and Electronics Engineers,2018:176-181.
[13]	龚钊宇.基于人工神经网络的民航发动机故障诊断技术[D].成都:电子科技大学,2017.
[14]	WISKOTT L,BERKES P,FRANZIUS M,et al.（2011） Slow feature analysis[EB/OL].[2020-09-05].http:∥scholarpedia.org/article/Slow_feature_analysis.
[15]	马奎俊,韩彦军,陶卿,等.基于核的慢特征分析算法[J].模式识别与人工智能,2011,24（2）:153-159.
[16]	COURANT R,HILBERT D.Methods of mathematical physics:partial differential equations[M].New Jersey,USA：John Wiley and Sons,2008.
[17]	郑慢慢.基于支持向量机的混合核函数研究[D].上海：华东师范大学,2019.
[18]	卢依容.基于核慢特征分析算法的故障检测与诊断[D].上海:上海交通大学,2015.
[19]	HARRINGTON P.机器学习实战[M].李锐,李鹏,曲亚东,译.北京：人民邮电出版社,2013.
[20]	ANKERST M,BREUNIG M M,KRIEGEL H,et al.OPTICS:ordering points to identify the clustering structure[J].International Conference on Management of Data,1999,28（2）:49-60.
[21]	SCHUBERT E,SANDER J,ESTER M,et al.DBSCAN revisited:why and how you should （still） use DBSCAN[J].ACM Transactions on Database Systems （TODS）,2017,42（3）:1-21.
[22]	WOLD S,ESBENSEN K,GELADI P.Principal component analysis[J].Chemometrics and Intelligent Laboratory Systems,1987,2（1/3）:37-52.
[23]	GOODFELLOW I,BENGIO Y,COURVILLE A.Deep learning[M].Cambridge,US:MIT Press,2016.
[24]	COMON P.Independent component analysis,a new concept?[J].Signal Processing,1994,36（3）:287-314.
[25]	ROWEIS S T,SAUL L K.Nonlinear dimensionality reduction by locally linear embedding[J].Science,2000,290（5500）:2323-2326.
[26]	CALI?SKI T,HARABASZ J.A dendrite method for cluster analysis[J].Communications in Statistics-theory and Methods,1974,3（1）:1-27.
[27]	ROUSSEEUW P J.Silhouettes:a graphical aid to the interpretation and validation of cluster analysis[J].Journal of Computational and Applied Mathematics,1987,20:53-65.
[28]	DAVIES D L,BOULDIN D W.A cluster separation measure[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1979（2）:224-227.
[29]	DUNN J C.A fuzzy relative of the ISODATA process and its use in detecting compact well-separated clusters[J].Journal of Statistical Computation and Simulation,1973,91（13）:2699-2716.
[30]	RODRIGUEZ A,LAIO A.Clustering by fast search and find of density peaks[J].Science,2014,344（6191）:1492-1496.