基于动态软聚类的航空电子部件LMKELM诊断模型

doi:10.12305/j.issn.1001-506X.2021.03.06

摘要/Abstract

摘要：

为提高小样本条件下航空电子设备模块级故障诊断精度, 基于动态软聚类的自适应特点与局部多核学习(local multiple kernel learning, LMKL)的局部特征表达能力, 提出一种新的局部多核超限学习机(local multiple kernel extreme learning machine, LMKELM)诊断模型。通过引入局部密度的概念进行自适应确定聚类数目, 并结合模糊C均值聚类对样本进行划分, 在充分体现类内多样性的同时, 约减了计算复杂度, 实现对样本的动态软聚类。通过构造选通函数解决局部权重二次非凸问题, 融合近似得到的局部权重与隶属度信息, 实现对测试样本的故障诊断。将该模型应用于某型机旋转变压器激励发生电路, 实验结果表明, 相比于4种前沿的多核学习方法, 该算法在漏警率、虚警率方面表现优异, 选用的M₁与M₂选通函数分别将诊断精度平均值提升2.78%和4.37%。

关键词: 故障诊断, 局部密度, 模糊C算法, 局部多核学习, 选通模型

Abstract:

To improve the module-level fault diagnosis rate for avionics in a small sample size, based on the adaptive feature of dynamic fuzzy clustering and the local feature of local multiple kernel learning (LMKL), a local multiple kernel extreme learning machine (LMKELM) model for avionics is proposed. The model confirms the number of clusters adaptively by introducing the concept of local density. Local density is combined with the fuzzy C-means algorithm to realize dynamic fuzzy clustering, which well reflects the diversity between clusters and reduces calculation. A gating model is constructed to solve the non-convex quadratic problem of local weight. Fault diagnosis of test samples is realized by fusing information of local weight and membership. The proposed model is applied into a certain type of rotary transformer excitation generating circuit. Compared with the four fashionable multiple kernel learning methods, the proposed model performs better in terms of false alarm and missing alarm rate. The experimental result shows that the applied M₁ and M₂ gating models enhance the average accuracy by 2.78% and 4.37% respectively.

Key words: fault diagnosis, local density, fuzzy C-means algorithm, local multiple kernel learning, gating model

中图分类号:

V243.1

戴金玲, 许爱强. 基于动态软聚类的航空电子部件LMKELM诊断模型[J]. 系统工程与电子技术, 2021, 43(3): 637-646.

Jinling DAI, Aiqiang XU. Local multiple kernel extreme learning machine fault diagnosis model with dynamic fuzzy clustering for avionics[J]. Systems Engineering and Electronics, 2021, 43(3): 637-646.

图/表 14

图1

图2

图3

图4

表1

图5

图6

图7

表2

图8

图9

表3

表4

表5

参考文献 31

1	ZHAO R , YAN R Q , CHEN Z H , et al. Deep learning and its applications to machine health monitoring[J]. Mechanical Systems and Signal Processing, 2019, 115, 213- 237. doi: 10.1016/j.ymssp.2018.05.050
2	HAN J H , CHOI D J , HONG S K . Study on development of deep learning fault diagnosis algorithm considering induction motor speed and loadcondition[J]. Transactions of the Korean Institute of Electrical Engineers, 2019, 68 (3): 423- 429. doi: 10.5370/KIEE.2019.68.3.423
3	TANG S H , SHEN C Q , WANG D , et al. Adaptive deep feature learning network with Nesterov momentum and its application to rotating machinery fault diagnosis[J]. Neurocomputing, 2018, 305 (AUG.30): 1- 14.
4	HAO X , ZHANG G G , MA S . Deep learning[J]. International Journal of Semantic Computing, 2016, 10 (3): 417- 439. doi: 10.1142/S1793351X16500045
5	TRIPOLITI E E , ZERVAKIS M , FOTIADIS D I . Multiple kernel learning algorithms and their use in biomedical informatics[M]. Springer International Publishing, 2016.
6	BIYUN A S , JUN L B , FU X A , et al. Multilayer deep features with multiple kernel learning for action recognition[J]. Neurocomputing, 2020, 399, 65- 74. doi: 10.1016/j.neucom.2020.02.096
7	刘敏, 张英堂, 李志宁, 等. 基于增量稀疏核极限学习机的柴油机故障在线诊断[J]. 上海交通大学学报, 2019, 53 (2): 95- 102.
	LIU M , ZHANG Y T , LI Z N , et al. Diesel engine fault online diagnosis method based on incremental sparse kernel extreme learning machine[J]. Journal of Shanghai Jiaotong University, 2019, 53 (2): 217- 224.
8	徐睿, 梁循, 齐金山, 等.极限学习机前沿进展与趋势[J/OL].[2020-03-29]. http://kns.cnki.net/kcms/detail/11.1826.TP.20190301.1029.002.html.
	XU R, LIANG X, QI J S, et al. Advances and trends in extreme learning machine[J/OL]. http://kns.cnki.net/kcms/detail/11.1826.TP.20190301.1029.002.html.
9	RAKOTOMAMONJY A , BACH F R , CANU S , et al. SimpleMKL[J]. Journal of Machine Learning Research, 2008, 9, 2491- 2521.
10	LIU X W , WANG L , HUANG G B , et al. Multiple kernel extreme learning machine[J]. Neurocomputing, 2015, 149 (pt, a): 253- 264.
11	WU D H , WANG Z L , CHEN Y , et al. Mixed-kernel based weighted extreme learning machine for inertial sensor basedhuman activity recognition with imbalanced dataset[J]. Neurocomputing, 2016, 190, 35- 49. doi: 10.1016/j.neucom.2015.11.095
12	LI X D , MAO W J , JIANG W . Multiple kernel learning based extreme learning machine for classification design[J]. Neural Computing and Applications, 2016, 27 (1): 175- 184. doi: 10.1007/s00521-014-1709-7
13	XUE J M , LIU Q , LI M M , et al. Incremental multiple kernel extreme learning machine and its application in Robo-advisors[J]. Soft Computing, 2018, 22 (11): 1- 11.
14	GÖNEN M , ALPAYDIN E . Multiple kernel learning algorithms[J]. Journal of Machine Learning Research, 2011, 12, 2211- 2268.
15	YE F M , ZHANG Z B , CHAKRABARTY K , et al. Board-level functional fault diagnosis using multi-kernel support vector machines and incremental learning[J]. IEEE Trans.on Computer-aided Design of Integrated Circuits and Systems, 2014, 33 (2): 279- 290. doi: 10.1109/TCAD.2013.2287184
16	ZHANG C L , HE Y G , YUAN L F , et al. A novel approach for diagnosis of analog circuit fault by using GMKL-SVM and PSO[J]. Journal of Electronic Testing, 2016, 32, 531- 540. doi: 10.1007/s10836-016-5616-y
17	LI Y X , REN C Q , BO J Y , et al. The application of GMKL algorithm to faultdiagnosis of local area network[J]. Journal of Networks, 2014, 9 (3): 747- 753.
18	GÖNEN M , ALPAYDIN E . Localized algorithms for multiple kernel learning[J]. Pattern Recognition, 2013, 46, 795- 807. doi: 10.1016/j.patcog.2012.09.002
19	WANG X M , HUANG Z X , DU Y J . Improving localized multiple kernel learning via radius-margin bound[J]. Mathematical Problems in Engineering, 2017, 2017 (1): 4579214.
20	HAN Y N , LIU G Z . Probability-confidence-kernel-based localized multiple kernel learning with L_p norm[J]. IEEE Trans.on Systems, Man and Cybernetics—Part B: Cybernetics, 2012, 42 (3): 827- 837. doi: 10.1109/TSMCB.2011.2179291
21	HAN Y N , YANG K D , LIU G Z . L_p norm localized multiple kernel learning via semi-definite programming[J]. IEEE Signal Processing Letters, 2012, 19 (10): 688- 691. doi: 10.1109/LSP.2012.2212431
22	HAN Y N , YANG K D , MA Y L , et al. Localized multiple kernel learning via sample-wise alternating optimization[J]. IEEE Trans.on Cybernetics, 2014, 44 (1): 137- 147. doi: 10.1109/TCYB.2013.2248710
23	LIU X W, WANG L, ZHANG J, et al. Sample-adaptive multiple kernel learning[C]//Proc.of the 28th AIAA Conference on Artificial Intelligence, 2014: 1975-1981.
24	CAO Z, PRINCIPE J C, OUYANG B. Group feature selection in image classification with multiple kernel learning[C] //Proc.of the IEEE International Joint Conference on Neural Networks, 2015.
25	YANG J J , TIAN Y H , LI Y N , et al. Group-sensitive multiple kernel learning for object recognition[J]. IEEE Trans.on Image Processing, 2012, 21 (5): 2838- 2852. doi: 10.1109/TIP.2012.2183139
26	张伟, 许爱强, 平殿发. 融合K近邻信息的局部三层交替优化MK诊断模型[J]. 仪器仪表学报, 2018, 39 (5): 123- 131.
	ZHANG W , XU A Q , PING D F . Multi-kernel fault diagnosis model with K nearest neighbor information based on localized three-step alternating optimization[J]. Chinese Journal of Scientific Instrument, 2018, 39 (5): 123- 131.
27	HAN Y N , YANG K D , YANG Y X , et al. Localized multiple kernel learning with dynamical clustering and matrix regularization[J]. IEEE Trans.on Neural Networks and Learning Systems, 2018, 29 (2): 486- 499.
28	张伟, 许爱强, 平殿发, 等. 基于近邻传播聚类的航空电子部件LMK诊断模型[J]. 北京航空航天大学学报, 2018, 44 (8): 1693- 1704.
	ZHANG W , XU A Q , PING D F , et al. Localized multi-kernel diagnosis model for avionics based on affinity propagation clustering[J]. Journal of Beijing University of Aeronautics and Astronautics, 2018, 44 (8): 1693- 1704.
29	朱敏, 许爱强, 李睿峰, 等. 基于隶属度和LMK-ELM的航空电子部件诊断方法[J]. 航空学报, 2019, 40 (12): 207- 219.
	ZHU M , XU A Q , LI R F , et al. Diagnosis method for avionics based on membership and LMK-ELM[J]. Acta Aeronautica Et Astronautica Sinica, 2019, 40 (12): 207- 219.
30	MIN R , PEI Y L , ZHI H W , et al. A self-adaptive fuzzy C-means algorithm for determining the optimal number of clusters[J]. Computational Intelligence and Neuroscience, 2016, 1- 12.
31	李米娜, 朱玉全, 陈耿, 等. 一种基于局部密度的核K-means算法[J]. 计算机应用研究, 2011, 28 (1): 78- 80. doi: 10.3969/j.issn.1001-3695.2011.01.020
	LI M N , ZHU Y Q , CHEN G , et al. Kernel K-means cluster-ing algorithm based on local density[J]. Computer Application and Rearch, 2011, 28 (1): 78- 80. doi: 10.3969/j.issn.1001-3695.2011.01.020

方法	分类精度	F1	G-mean
SimpleMKL^[9]	0.856 9±0.012 1	0.862 1±0.012 4	0.858 8±0.011 9
GMKL-SVM^[17]	0.865 5±0.007 3	0.870 1±0.011 8	0.871 1±0.004 8
l₁-FCLMKELM^[29]	0.869 9±0.009 7	0.881 2±0.011 1	0.876 7±0.010 1
M₂-LCMKELM^[28]	0.876 0±0.119 0	0.884 8±0.005 0	0.879 4±0.010 3
M₁-DFCMKELM	0.880 4±0.016 5	0.884 5±0.012 1	0.883 0±0.014 3
M₂-DFCMKELM	0.893 8±0.015 9	0.891 8±0.015 8	0.8950±0.015 3

样本序号	聚类中心
样本序号	1	2	3	4
1	0.456 384	0.751 081	0.706 339	0.150 229
2	0.230 475	0.106 382	0.190 408	0.097 19
3	0.313 141	0.142 537	0.103 253	0.752 581
4	0.123 149	0.193 747	0.122 144	0.243 813
				
63	0.837 557	0.632 522	0.812 26	0.274 315

算法	漏警率	虚警率	训练准确率	故障准确率
SimpleMKL	6.349 2	0	100	88.888 9
GMKL-SVM	0	3.174 6	100	92.063 5
l₁-FCLMKELM	1.587 3	0	100	95.238 1
M₂-LCMKELM	0	0	100	93.650 8
M₁-DFCMKELM	0	0	100	95.238 1
M₂-DFCMKELM	0	0	100	96.825 4

算法	F1值	G-mean
SimpleMKL	0.901 4	0.892 3
GMKL-SVM	0.924 6	0.915 8
l₁-FCLMKELM	0.953 3	0.950 1
M₂-LCMKELM	0.942 1	0.934 7
M₁-DFCMKELM	0.957 5	0.950 3
M₂-DFCMKELM	0.970 6	0.964 8

算法	训练时间	测试时间
SimpleMKL	0.665 2±0.020 1	0.169 5±0.003 5
GMKL-SVM	0.756 4±0.018 6	0.164 8±0.002 4
l₁-FCLMKELM	1.943 2±0.047 3	0.138 3±0.011 4
M₂-LCMKELM	1.831 9±0.083 9	0.138 4±0.005 8
M₁-DFCMKELM	1.864 7±0.164 6	0.127 3±0.012 9
M₂-DFCMKELM	1.756 1±0.223 1	0.138 2±0.011 8