应用深度核极限学习机的航空发动机部件故障诊断

推进技术 ›› 2017, Vol. 38 ›› Issue (11): 2613-2621.

应用深度核极限学习机的航空发动机部件故障诊断

逄珊¹,杨欣毅²,张勇²,韦祥²

鲁东大学信息与电气工程学院，山东烟台 264025,海军航空大学航空基础学院，山东烟台 264001,海军航空大学航空基础学院，山东烟台 264001,海军航空大学航空基础学院，山东烟台 264001

发布日期:2021-08-15
作者简介:逄珊，女，硕士，讲师，研究领域为模式识别理论与应用。
基金资助:
山东省自然科学基金(ZR2016FQ19)。

Application of Deep Kernel Extreme Learning Machine in Aero Engine Components Fault Diagnosis

College of Information and Electrical Engineering，Ludong University，Yantai 264025，China,Aeronautical Foundation College，Naval Aeronautical University，Yantai 264001，China,Aeronautical Foundation College，Naval Aeronautical University，Yantai 264001，China and Aeronautical Foundation College，Naval Aeronautical University，Yantai 264001，China

Published:2021-08-15

摘要/Abstract

摘要： 运用传统单隐层的神经网络进行航空发动机部件故障诊断识别受其浅层结构影响，精度不高，而用深度置信网络(Deep belief network，DBN)等深度学习方法则存在耗时、参数训练复杂的问题。为解决现有的基于数据驱动的发动机部件故障诊断方法的不足，提高诊断精度，缩短训练时间，将核方法和多层极限学习机(Multilayer extreme learning machine，M-ELM)相结合，提出一种深度核极限学习机(Deep kernel extreme learning machine，DK-ELM)。算法首先利用深度网络结构对输入数据进行逐层的特征提取，抽象得到的特征通过核函数实现高维空间映射分类。这些措施有利于提高算法的分类精度和泛化性能，在训练速度上较深度学习也有明显的提高。将该算法与深度学习和其他极限学习机算法进行综合比较研究，结果表明:基于DK-ELM的诊断方法有效、可靠，便于实现，为航空发动机部件故障诊断提供一个更为优秀实用的工具。

关键词: 涡扇发动机；部件；故障诊断；深度神经网络；极限学习机；核方法

Abstract: Traditional aero engine components fault diagnosis methods based on single hidden layer neural network are limited in test accuracy due to their shallow network structure. While deep learning methods，such as deep belief network (DBN)，have the problems as time-consuming and difficult to train parameters. In order to solve the problems of existing data-driven aero engine components fault diagnosis methods and improve test accuracy at the same time shorten training time，we proposed a new deep kernel extreme learning machine (DK-ELM) which combines kernel method with multilayer extreme learning machine. It first abstracts the original input data layer by layer gradually. Then，the abstracted feature was mapped to higher dimensional space through kernel function and is classified. These measures help to improve the algorithm in its testing accuracy and generalization performance. And it has more fast training speed than deep learning methods. We have compared the proposed method with other deep learning methods and ELM based methods on some benchmark classification data sets and engine components fault data set. Results show that the DK-ELM based method is effective，reliable and easy to implement，thus it proves to be a good and functional tool for engine components fault diagnosis.

Key words: Gas turbine fan engine；Component；Defect diagnosis；Deep neural network；Extreme learning machine；Kernel method

逄珊,杨欣毅,张勇,韦祥. 应用深度核极限学习机的航空发动机部件故障诊断[J]. 推进技术, 2017, 38(11): 2613-2621.

PANG Shan¹,YANG Xin-yi²,ZHANG Yong²,WEI Xiang². Application of Deep Kernel Extreme Learning Machine in Aero Engine Components Fault Diagnosis[J]. Journal of Propulsion Technology, 2017, 38(11): 2613-2621.

参考文献

[1] Xinyi Yang, Wei Shen, Shan Pan. A Novel Gas Turbine Engine Health Status Estimation Method Using Quantum-Behaved Particle Swarm Optimization[J].Mathematical Problems in Engineering, 2014, (3): 1-11.
[2] 邓明, 金业壮. 航空发动机故障诊断[M]. 北京: 北京航空航天大学出版社, 2012: 102-105.
[3] Tan H S. Fourier Neural Networks and Generalized Single Hidden Layer Networks in Aircraft Engine Fault Diagnostics[J]. Journal of Engineering for Gas Turbines and Power, 2006, 128(4): 773-782.
[4] Romessis C, Mathioudakis K. Bayesian Network Approach for Gas Path Fault Diagnosis[J]. Journal of Engineering for Gas Turbines and Power, 2006, 128(1): 64-72.
[5] Sang-Myeong Lee, Tae-Seong Roh, Dong-Whan Choi. Defect Diagnostics of SUAV Gas Turbine Engine Using Hybrid SVM-Artificial Neural Network Method[J]. Journal of Mechanical Science and Technology, 2009, 23(2): 559-568.
[6] Sang-Myeong Lee, Won-Jun Choi, Tae-Seong Roh. A Study on Separate Learning Algorithm Using Support Vector Machine for Defect Diagnostics of Gas Turbine Engine[J]. Journal of Mechanical Science and Technology, 2008, 22(12): 2489-2497.
[7] Sadough Vanini Z N, Khorasani K, Meskin N. Defect Detection and Isolation of a Dual Spool Gas Turbine Engine Using Dynamic Neural Networks and Multiple Model Approach[J]. Information Sciences, 2014 , 259(2): 234-251.
[8] 徐启华, 耿帅, 师军. 基于大规模训练集SVM的发动机故障诊断[J]. 航空动力学报, 2011, 26(12): 2841-2848.
[9] 徐启华, 师军, 耿帅. 应用快速多分类SVM的航空发动机故障诊断方法[J]. 推进技术, 2012, 33(6):961-967. (XU Qi-hua, SHI Jun, GENG Shuai. Aero-Engine Fault Diagnosis by a New Fast Multi-Class Support Vector Algorithm[J]. Journal of Propulsion Technology, 2012, 33(6): 961-967.)
[10] 胡金海, 谢寿生, 杨帆, 等. 基于支持向量机的组合分类方法及应用[J]. 推进技术, 2007, 28(6):669-673. (HU Jin-hai, XIE Shou-sheng, YANG Fan, et al. Ensemble of Classification Methods Based on SVM and Its Application in Diagnosis[J]. Journal of Propulsion Technology, 2007, 28(6): 669-673.)
[11] Huang G B, Ding X, Zhou H. Optimization Method Based Extreme Learning Machine for Classification[J]. Neurocomputing, 2010, 74(1-3): 155-163.
[12] 张英堂, 马超尹, 尹刚, 等. 基于多极限学习机集成的柴油机故障诊断方法研究[J]. 车用发动机, 2012, 6: 85-89.
[13] Guang-Bin Huang, Qin-Yu Zhu, Chee-Kheong Siew. Extreme Learning Machine: Theory and Applications[J]. Neurocomputing, 2006, 70(1-3): 489-501.
[14] Huang G-Bin, Zhou H, Ding X, et al. Extreme Learning Machine for Regression and Multiclass Classification[J]. IEEE Trans.Systems, Man, and Cybernetics, 2012, 42(2): 513-529.
[15] 李业波, 李秋红, 黄向华, 等. 航空发动机气路部件故障融合诊断方法研究[J]. 航空学报, 2014, 35(6):1612-1622.
[16] 李业波, 李秋红, 黄向华, 等. 航空发动机传感器故障与部件故障诊断技术[J]. 北京航空航天大学学报, 2013, 39(9): 1174-1180.
[17] Xinyi Yang, Shan Pang, Wei Shen, et al. Aero Engine Fault Diagnosis Using an Optimized Extreme Learning Machine[J]. International Journal of Aerospace Engineering, 2016, (1): 1-10.
[18] Hinton G, Salakhutdinov R. Reducing the Dimensionality of Data with Neural Networks[J]. Science, 2006, 313(504).
[19] Vincent P, Larochelle H, Bengio Y, et al. Extracting and Composing Robust Features with Denoising Autoencoders[C]. Helsinki: Proceedings of the International Conference on Machine Learning, 2008: 1096-1103.
[20] Hinton G E, Krizhevsky A, Wang S. Transforming Auto-Encoders[C]. Espoo: Proceedings of the International Conference on Artificial Neural Networks, 2011: 44-51.
[21] Schmidhuber J. Deep Learning in Neural Networks: An Overview[J]. Neural Networks, 2015, 61(1): 85-117.
[22] 余凯, 贾磊, 陈雨强, 等. 深度学习的昨天、今天和明天[J]. 计算机研究与发展, 2013, 50(9): 1799-1804.
[23] 焦李成, 杨淑媛, 刘芳, 等. 神经网络七十年:回顾与展望[J]. 计算机学报, 2016, 39(5): 1-21.
[24] Prasanna Tamilselvan, Pingfeng Wang. Failure Diagnosis Using Deep Belief Learning Based Health State Classification[J]. Reliability Engineering and System Safety, 2013, 115(1): 124-135.
[25] 李本威, 林学森, 杨欣毅. 深度置信网络在发动机部件性能退化故障诊断中的应用研究[J]. 推进技术, 2016, 37(11). (LI Ben-wei, LIN Xue-seng, YANG Xin-yi. Research on the Application of Deep Belief Network in Aero Engine Component Performance Deterioration Fault Diagnosis[J]. Journal of Propulsion Technology, 2016, 37(11).)
[26] 雷亚国, 贾峰, 周昕, 等. 基于深度学习理论的机械装备大数据健康监测方法[J]. 机械工程学报, 2015, 51(1): 49-56.
[27] Noel Lopes, Bernardete Ribeiro. Towards Adaptive Learning with Improved Convergence of Deep Belief Network on Graphics Processing Units[J]. Pattern Recognition, 2014, 47(1): 114-127.
[28] Liyanaarachchi Lekamalage Chamara Kasun, Hongming Zhou, Guang-Bin Huang. Representational Learning with ELMs for Big Data[J]. IEEE Intelligent Systems, 2013, 28(6): 31-34.
[29] Cortes C, Vapnik V. Support Vector Networks[J]. Machine Learning, 1995, 20(3): 273-297.
[30] Huang G B, Zhou H, Ding X, et al. Extreme Learning Machine for Regression and Multiclass Classification[J]. IEEE Trans.Syst.Man, Cybern.B, Cybern, 2012, 42(2): 513-529.
[31] Jiexiong Tang, Chenwei Deng, Huang G B. Extreme Learning Machine for Multilayer Perception[J]. IEEE Transactions of Neural Network and Learning Systems, 2015, 27(4): 809-821.(编辑:梅瑛)　　　　　　　　　　　　　*　收稿日期:2016-06-05；修订日期:2016-08-23。基金项目:山东省自然科学基金(ZR2016FQ19)。作者简介:逄珊,女,硕士,讲师,研究领域为模式识别理论与应用。E-mail: pangshanpp@163.com