基于张量Tucker分解的发动机故障诊断

中国机械工程

基于张量Tucker分解的发动机故障诊断

许小伟1,2;沈琪1;严运兵1,2;吴强1;张楠1

1.武汉科技大学汽车与交通工程学院，武汉， 430081
2.纯电动汽车动力系统设计与测试湖北省重点实验室，襄阳，441053

出版日期:2018-03-10 发布日期:2018-03-08
基金资助:
国家自然科学基金资助项目（51505345）；
电动汽车动力系统设计与测试湖北省重点实验室基金资助项目（HBUASEV2015F005）；
湖北省教育厅基金资助项目（Q20151105）
National Natural Science Foundation of China （No. 51505345）

Engine Fault Diagnosis Based on Tensor Tucker Decomposition

XU Xiaowei1,2;SHEN Qi1;YAN Yunbing1,2;WU Qiang1;ZHANG Nan1

1.School of Automobile and Traffic Engineering,Wuhan University of Science and Technology,Wuhan,430081
2.Hubei Key Laboratory of Power System Design and Test for Electrical Vehicle,Xiangyang,Hubei,441053

Online:2018-03-10 Published:2018-03-08
Supported by:
National Natural Science Foundation of China （No. 51505345）

摘要/Abstract

摘要： 传统的发动机故障诊断方法通常基于向量模式进行数据特征的提取，可能丢失数据之间的结构信息及破坏数据间相关性。针对此问题，提出了一种张量模式下提取发动机数据特征的方法，构建了“信号类别×曲轴转角×转速”的三阶张量形式的发动机状态样本，基于交替投影的思想，使用HOSVD-HOOI张量Tucker分解的联立求解算法，对数据特征进行提取。分别以不进行数据特征提取和基于张量Tucker分解进行数据特征提取两种情况，对发动机正常工作、单缸失火和轴系不对中三种状态下的实验数据进行处理，并分别采用网格参数优化法、遗传算法、粒子群算法对分类模型中的参数进行优化。以预测准确率和模型学习时间为评价指标进行对比分析，实验结果表明，基于张量Tucker分解的发动机数据特征提取及诊断方法预测准确率更高，分类模型学习时间更短。

关键词: 发动机, 故障诊断, 张量模式, Tucker分解

Abstract: The traditional engine fault diagnosis method based on vector mode to extract data features was a method which might lose the structure informations of the data and damage the correlation between the data. To solve this problem, the paper presented a method to extract engine data features in tensor mode, and the engine state samples of the three order tensor form of the “signal class×crank angle×rotation rate” were constructed.Based on alternating projection theory, the HOSVD-HOOI tensor Tucker decomposition of the simultaneous solution algorithm was used to extract data features.The data were processed without feature extractions and with the method based on the tensor Tucker decomposition respectively,and the experimental data were processed under three kinds of state of normal operation, single cylinder misfire and shaft misalignment .The parameters of the classification model was optimized by using the grid parameter optimization method, genetic algorithm and particle swarm optimization respectively.Then the prediction accuracy and model learning time were compared and analyzed as evaluating indicator.The experimental results show that the prediction accuracy of engine data feature extraction and diagnosis method based on tensor Tucker decomposition is higher, and the learning time of classification model is shorter.

Key words: engine, fault diagnosis, tensor mode, Tucker decomposition

中图分类号:

TK411

许小伟1,2;沈琪1;严运兵1,2;吴强1;张楠1. 基于张量Tucker分解的发动机故障诊断[J]. 中国机械工程.

XU Xiaowei1,2;SHEN Qi1;YAN Yunbing1,2;WU Qiang1;ZHANG Nan1. Engine Fault Diagnosis Based on Tensor Tucker Decomposition[J]. China Mechanical Engineering.

参考文献

[1]张宇飞, 么子云, 唐松林,等. 一种基于主成分分析和支持向量机的发动机故障诊断方法[J]. 中国机械工程, 2016, 27(24): 3307-3311.
ZHANG Yufei,YAO Ziyun,TANG Songlin, et al. An Engine Fault Diagnosis Method Based on PCL and SVM[J]. China Mechanical Engineering, 2016,27(24): 3307-3311.
[2]HOU C, NIE F, ZHANG C, et al. Multiple Rank Multi-linear SVM for Matrix Data Classification[J]. Pattern Recognition, 2013, 47(1): 454-469.
[3]杨兵. 基于张量数据的机器学习方法研究与应用[D]. 北京：中国农业大学, 2014.
YANG Bing. Research and Application of Machine Learning Algorithm Based Tensor Representation [D]. Beijing:China Agricultural University, 2014.
[4]KARAMI A, YAZDI M, MERCIER G. Compression of Hyperspectral Images Using Discerete Wavelet Transform and Tucker Decomposition[J]. IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing, 2012, 5(2): 444-450.
[5]BENETOS E, KOTROPOULOS C. Non-negative Tensor Factorization Applied to Music Genre Classification[J]. IEEE Transactions on Audio Speech & Language Processing, 2010, 18(8): 1955-1967.
[6]ZHANG Q, WANG Y, LEVINE M D, et al. Multisensor Video Fusion Based on Higher Order Singular Value Decomposition[J]. Information Fusion, 2015, 24(C): 54-71.
[7]BARANYI P. Numerical Reconstruction of the HOSVD Based Canonical Form of Polytopic Dynamic Models[C]// International Symposium on Computational Intelligence & Intelligent Informatics. New York, 2007: 111-116.
[8]SHEEHAN B N, SAAD Y. Higher Order Orthogonal Iteration of Tensors (HOOI) and Its Relation to PCA and GLRAM[C]//SIAM International Conference on Data Mining. Minneapolis, Minnesota, USA, 2007:196-199.
[9]HACKBUSCH W. Tensor Spaces and Numerical Tensor Calculus[M].Heidelberg:Springer, 2012.
[10]KOLDA T G, BADER B W. Tensor Decompositions and Applications[J]. SIAM Review, 2009, 51(3): 455-500.
[11]CARROLL J D, CHANG J J. Analysis of Individual Differences in Multidimensional Scaling via an n-way Generalization of “Eckart-Young” Decomposition[J]. Psychometrika, 1970, 35(3): 283-319.
[12]刘亚楠.基于图和低秩表示的张量分解方法及应用研究[D]. 合肥：安徽大学, 2014.
LIU Yanan. Tensor Decomposition and Its Applications Based on Graph and Low Rank Representation[D]. Hefei:Anhui University,2014.
[13]SUN J, TAO D, PAPADIMITRIOU S, et al. Incremental Tensor Analysis: Theory and Applications[J]. ACM Transactions on Knowledge Discovery from Data, 2008, 2(3): 651-678.
[14]王健峰. 基于改进网格搜索法SVM参数优化的说话人识别研究[D]. 哈尔滨：哈尔滨工程大学, 2012.
WANG Jianfeng. Speaker Recognition Based on Improvedgrid Search Algorithm for SVM Parameter Optimization[D]. Harbin: Harbin Engineering University, 2012.
[15]汪松泉. 遗传算法在组合优化中的应用研究[D]. 合肥：安徽大学, 2010.
WANG Songquan. Application of Genetic Algorithm in Combinatorial Optimization[D]. Hefei:Anhui University,2010.
[16]张丽平. 粒子群优化算法的理论及实践[D]. 杭州：浙江大学, 2005.
ZHANG Liping. Theory and Practice of Particle Swarmoptimization Algorithm[D]. Hangzhou:Zhejiang University, 2005.