基于RS-CART决策树的航空发动机小样本故障诊断

庞梦洋; 索中英; 郑万泽

doi:10.13224/j.cnki.jasp.2020.07.024

基于RS-CART决策树的航空发动机小样本故障诊断

doi: 10.13224/j.cnki.jasp.2020.07.024

基金项目: 国家自然科学基金（61772021）

计量
- 文章访问数: 292
- HTML浏览量: 10
- PDF量: 504
- 被引次数: 0
出版历程
- 收稿日期: 2019-12-23
- 刊出日期: 2020-07-28

Small sample fault diagnosis of aeroengine based on RS-CART decision tree

1.
Department of Basic Sciences,Air force Engineering University,Xi’an 710051,China
2.
Academic Research Office,Air force Engineering University,Xi‘an 710051,China

摘要

摘要: 针对CART(classification and regression tree)分类决策树构建过程中由于小样本集特征维数高及噪声等造成的过拟合问题,在CART决策树算法训练过程中引入基于互信息的粗糙集(rough sets,RS)属性约简,考虑信息熵与基尼(GINI)系数刻画样本集“纯净度”的相似关系,对历史故障数据进行属性约简,降低属性维度以优化训练集,在此基础上构建分类决策树,可视化输出规则。实验结果表明:将改进的CART决策树算法应用于某型航空发动机油液故障诊断,提取的规则可解释性强,能够减小冗余属性及噪声对决策的影响,与常用故障诊断算法相比,该模型的诊断准确率提升20%左右,AUC(area under curve)值高达92%,可以有效处理高维离散型航空发动机小样本故障问题。
- 故障诊断 /
- 属性约简 /
- 分类与回归树 /
- GINI系数 /
- 规则提取
Abstract: In view of the over-fitting problems caused by the high dimension of feature attributes of small sample sets and noise data, etc, during the construction of CART (classification and regression tree) classification tree, the rough set (RS) attribute reduction based on mutual information was introduced in the process of CART decision tree algorithm training. The similarity between information entropy and GINI coefficients to characterize the “purity” of the sample set was considered, attribute reduction on historical fault data was performed to reduce the attribute dimensions and optimize the training sets.On this basis,a classification decision tree was constructed and the output rules were visualized.Result showed that,when the improved CART decision tree algorithm was applied to the fault diagnosis of a certain type of aeroengine oil,the extracted rules were highly interpretable and can reduce the influence of redundant attributes and noise on decision.Compared with the common fault diagnosis algorithms,the diagnostic accuracy of the model was improved by about 20%,and AUC (area under curve) value was as high as 92%, enabling to effectively deal with the problem of small sample of high-dimensional discrete aeroengine faults.
- fault diagnosis /
- attribute reduction /
- classification and regression tree /
- ,GINI coefficient /
- rule extraction

HTML

参考文献(26)

[1]	于劲松,沈琳,唐荻音,等.基于贝叶斯网络的故障诊断系统性能评价［J］.北京航空航天大学学报,2016,42(1):35-40. YU Jinsong,SHEN Lin,TANG Diyin,et al.Performance evaluation of fault diagnosis system based on Bayesian network［J］.Journal of Beijing University of Aeronautics and Astronautics,2016,42(1):35-40.(in Chinese)
[2]	曹惠玲,高升,薛鹏.基于多分类AdaBoost的航空发动机故障诊断［J］.北京航空航天大学学报,2018,44(9):1818-1825. CAO Huiling,GAO Sheng,XUE Peng.Aeroengine fault diagnosis based on multi-classification AdaBoost［J］.Journal of Beijing University of Aeronautics and Astronautics,2018,44(9):1818-1825.(in Chinese)
[3]	赵世荣,黄向华.应用神经网络信息融合诊断航空发动机故障［J］.航空动力学报,2008,23(1):163-168. ZHAO Shirong,HUANG Xianghua.Fault diagnosis for aeroengine gas path components based on neural network multisensor data fusion［J］.Journal of Aerospace Power,2008,23(1):163-168.(in Chinese)
[4]	车畅畅,王华伟,倪晓梅,等.基于深度学习的航空发动机故障融合诊断［J］.北京航空航天大学学报,2018,44(3):621-628. CHE Changchang,WANG Huawei,NI Xiaomei,et al.Fault fusion diagnosis of aero-engine based on deep learning［J］.Journal of Beijing University of Aeronautics and Astronautics,2018,44(3):621-628.(in Chinese)
[5]	程嗣怡,索中英,吴华,等.基于协调近似表示空间的航空发动机故障诊断［J］.航空动力学报,2009,24(7):1644-1648. CHENG Siyi,SUO Zhongying,WU hua,et al.Aeroengine fault diagnosis based on consistent-approximative denoted space［J］.Journal of Aerospace Power,2009,24(7):1644-1648.(in Chinese)
[6]	BENKERCHA R,MOULAHOUM S.Fault detection and diagnosis based on C4.5 decision tree algorithm for grid connected PV system［J］.Solar Energy,2018,173(7):610-634.
[7]	范庚,马登武,张继军,等.基于决策树和相关向量机的智能故障诊断方法［J］.计算机工程与应用,2013,49(14):267-270. FAN Geng,MA Dengwu,ZHANG Jijun,et al.Intel-ligent fault diagnosis method based on decision tree and RVM［J］.Computer Engineering and Applications,2013,49(14):267-270.(in Chinese)
[8]	张耀鹏.基于决策树算法的扼流适配变压器故障诊断系统研究与设计［D］.北京：北京交通大学,2018. ZHANG Yaopeng.Research and design of fault diagnosis system of impedance match bond based on decision tree algorithms［D］.Beijing:Beijing Jiaotong University,2018.(in Chinese)
[9]	朱文博.基于改进决策树的轨道电路故障诊断方法研究［D］.成都:西南交通大学,2017. ZHU Wenbo.Research on fault diagnosis for track circuits based on improved decision tree［D］.Chengdu:Southwest Jiaotong University,2017.(in Chinese)
[10]	TRONCOSO A,SALCEDO-SANZ S,CASANOVA-MATEO C,et al.Local models-based regression trees for very shortterm wind speed prediction［J］.Renewable Energy,2015,81(3):589-598.
[11]	SRIVASTAVA R,TIWARI A N,GIRI V K.Solar radiation forecasting using MARS,CART,M5,and random forest model:a case study for India［J］.Heliyon,2019,5(10):1-14.
[12]	赵荣珍,王雪冬,邓林峰.基于PCA-KLFDA的小样本故障数据集降维方法［J］.华中科技大学学报(自然科学版),2015,43(12):12-16. ZHAO Rongzhen,WANG Xuedong,DENG Linfeng.Small sample size fault data recognition based on the principal component analysis and kernel local Fisher discriminant analysis［J］.Journal of Huazhong University of Science and Technology (Natural Science Edition),2015,43(12):12-16.(in Chinese)
[13]	PAWLAK Z.Rough sets［J］.International Journal of Computer and Information Sciences,1982,11(5):341-356.
[14]	于洪,王国胤,姚一豫.决策粗糙集理论研究现状与展望［J］.计算机学报,2015,38(8):1628-1639. YU Hong,WANG Guoyin,YAO Yiyu.Current research and future perspectives on decision-theoretic rough sets［J］.Chinese Journal of Computers,2015,38(8):1628-1639.(in Chinese)
[15]	张亮,宁芊.CART决策树的两种改进及应用［J］.计算机工程与设计,2015,36(5):1209-1213. ZHANG Liang,NING Qian.Two improvements on CART decision tree and its application［J］.Computer Engineering and Design,2015,36(5):1209-1213.(in Chinese)
[16]	张文修,吴伟志,梁吉业,等.粗糙集理论与方法［M］.北京:科学出版社,2001.
[17]	GONZALEZ-LOPEZ J,VENTURA S,CANO A.Distributed selection of continuous features in multilabel classification using mutual information［J］.IEEE Transactions on Neural Networks and Learning Systems,2019,30(9):1-14.
[18]	项海飞.基于互信息粒度的相对约简的矩阵计算方法［J］.西南师范大学学报(自然科学版),2014,39(3):60-64. XIANG Haifei.Matrix computing method based on relative reduction of mutual information granularity［J］.Journal of Southwest China Normal University (Natural Science Edition),2014,39(3):60-64.(in Chinese)
[19]	王金杰,李炜.混合互信息和粒子群算法的多目标特征选择方法［J］.计算机科学与探索,2020,14(1):83-95. WANG Jinjie,LI Wei.Multi-objective feature selection method based on hybrid MI and PSO algorithm［J］.Journal of Frontiers of Computer Science and Technology,2020,14(1):83-95.(in Chinese)
[20]	苗夺谦,王珏.粗集理论中概念与运算的信息表示［J］.软件学报,1999,10(2):113-116. MIAO Duoqian,WANG Jue.Information representation of concepts and operations in rough set theory［J］.Journal of Software,1999,10(2):113-116.(in Chinese)
[21]	李航.统计学习方法［M］.北京:清华大学出版社,2012.
[22]	程华,李艳梅,罗谦,等.基于C4.5决策树方法的到港航班延误预测问题研究［J］.系统工程理论与实践,2014,34(增刊1):239-247. CHENG Hua,LI Yanmei,LUO Qian,et al.Study on flight delay with C4.5 decision tree based prediction method［J］.Systems Engineering:Theory and Practice,2014,34(Suppl.1):239-247.(in Chinese)
[23]	孙梦婷,魏海平,李星滢等.利用CART分类树分类检测交通拥堵点［EB/OL］.［2019-12-23］.https:∥doi.org/10.13203/j.whugis20190288.
[24]	李国和,王峰,郑阳,等.基于决策树生成及剪枝的数据集优化及其应用［J］.计算机工程与设计,2018,39(1):205-211. LI Guohe,WANG Feng,ZHENG Yang,et al.Optimization of data set and its application based on construction and pruning of decision tree［J］.Computer Engineering and Design,2018,39(1):205-211.(in Chinese)
[25]	黄锦静,陈岱,李梦天.基于粗糙集的决策树在医疗诊断中的应用［J］.计算机技术与发展,2017,27(12):148-152. HUANG Jinjing,CHEN Dai,LI Mengtian.Application of decision tree based on rough set in medical diagnosis［J］.Computer Technology and Development,2017,27(12):148-152.(in Chinese)
[26]	索中英,朱林户,吴华,等.基于变精度粗糙集的航空发动机故障诊断［J］.航空动力学报,2008,23(10):1842-1846. SUO Zhongying,ZHU Linhu,WU Hua,et al.Aeroengine fault diagnosis based on variable precision rough sets［J］.Journal of Aerospace Power,2008,23(10):1842-1846.(in Chinese)