基于样本重采样的电路非平衡数据预处理方法

doi:10.3969/j.issn.1001-506X.2020.11.30

摘要/Abstract

摘要：

针对机载设备电子电路故障状态测试数据少、整体测试数据不均衡的问题,提出了一种基于样本重采样的数据预处理方法。首先，采用超限学习机对原始数据集进行训练以挑选出分类准确的样本。然后，对其中的少数类和多数类分别采用合成少数类过采样技术(synthetic minority oversampling technique, SMOTE)进行过采样和局部密度欠采样处理;并将错误分类的多数类样本作为干扰因素进行删除。通过以上两种手段可以均衡数据集,并控制数据规模防止过拟合,提高对故障样本的检测率。实测数据处理结果表明,相比于其他重采样算法,所提算法整体效果优良且稳定,对电子电路故障诊断具有一定的应用价值。

关键词: 电子电路, 非平衡数据, 重采样, 局部密度, 分类

Abstract:

In order to solve the deficiency of fault state data and imbalance of whole test data in airborne electronic circuit, a data preprocessing method based on sample resampling is proposed. Firstly, extreme learning machine is used to training the original data set to select the correct classified samples. Secondly, the synthetic minority oversampling technique (SMOTE) is used to oversampling and local density under-sampling respectively for the minority and majority of the correct classified samples. And the misclassified majority samples are deleted as interference factors. In this way, the data set can be equalized, and the data size can be controlled to prevent over-fitting, and the detection rate of fault samples can be improved. Compared with other data resampling methods, the test data processing results show that the proposed method has a good and stable overall effect, which has a certain application value for the fault diagnosis of electronic circuit.

Key words: electronic circui, imbalanced data, resample, local density, classification

中图分类号:

TP206.1

李睿峰, 许爱强, 孙伟超, 吴阳勇. 基于样本重采样的电路非平衡数据预处理方法[J]. 系统工程与电子技术, 2020, 42(11): 2654-2660.

Ruifeng LI, Aiqiang XU, Weichao SUN, Yangyong WU. Preprocessing method based on sample resampling for imbalanced data of electronic circuits[J]. Systems Engineering and Electronics, 2020, 42(11): 2654-2660.

图/表 8

图1

表1

表2

图2

图3

表3

表4

图4

参考文献 32

1	景博, 黄以锋, 张建业. 航空电子系统故障预测与健康管理技术现状与发展[J]. 空军工程大学学报(自然科学版), 2010, 11 (6): 1- 6.
	JING B , HUANG Y F , ZHANG J Y . Status and perspectives of prognostics and health management technology of avionics system[J]. Journal of Air Force Engineering University (Natural Science Edition), 2010, 11 (6): 1- 6.
2	HENAO H , CAPOLINO G A , FERNANDEZ-CABANAS M , et al. Trends in fault diagnosis for electrical machines:a review of diagnostic techniques[J]. IEEE Industrial Electronics Magazine, 2014, 8 (2): 31- 42. doi: 10.1109/MIE.2013.2287651
3	GAO M , HONG X , CHEN S , et al. A combined SMOTE and PSO based RBF classifier for two class imbalanced problems[J]. Neuro Computing, 2011, 74 (17): 3456- 3466.
4	BELLINO A , FASANA A , GARIBALDI L , et al. PCA-based detection of damage in time-varying systems[J]. Mechanical Systems & Signal Processing, 2010, 24 (7): 2250- 2260.
5	SHIVAKUMAR K , KUMAR K S V P , SUMANTH K . An artificial immune system approach for the fault detection and diagnosis of a DC machine[J]. Journal of Medical Virology, 2014, 42 (4): 374- 379.
6	NA S G , YANG I B , HEO H . Abnormality detection via SVDD technique of motor-generator system in HEV[J]. International Journal of Automotive Technology, 2014, 15 (4): 637- 643. doi: 10.1007/s12239-014-0066-y
7	XIAO Y C , WANG H G , ZHANG L , et al. Two methods of selecting Gaussian kernel parameters for one-class SVM and their application to fault detection[J]. Knowledge Based Systems, 2014, 59 (3): 75- 84.
8	MIAO Z M, ZHAO L W, YUAN W W, et al. Multi-class imba-lanced learning implemented in network intrusion detection[C]//Proc.of the International Conference on Computer Science and Service System, 2011: 1395-1407.
9	SMAILOVIĆ J , GRČAR M , LAVRAČ N , et al. Stream-based active learning for sentiment analysis in the financial domain[J]. Information Sciences, 2014, 285 (C): 181- 203.
10	LIU Y Q, WANG C, ZHANG L. Decision tree based predictive models for breast cancer survivability on imbalanced data[C]//Proc.of the 3rd International Conference on Bioinformatics and Biomedical Engineering, 2009: 312-315.
11	古平, 欧阳源遊. 基于混合采样的非平衡数据集分类研究[J]. 计算机应用研究, 2015, 32 (2): 379- 381.
	GU P , OUYANG Y Y . Classification research for unbalanced data based on mixed-sampling[J]. Applications Research of Computer, 2015, 32 (2): 379- 381.
12	YU H L , YANG X B , ZHENG S , et al. Active learning from imbalanced data: a solution of online weighted extreme learning machine[J]. IEEE Trans.on Neural Networks and Learning Systems, 2019, 30 (4): 1088- 1103. doi: 10.1109/TNNLS.2018.2855446
13	CHEN S , HE H B , GARCIA E A . RAMOBoost: ranked minority oversampling in boosting[J]. IEEE Trans.on Neural Networks/a Publication of the IEEE Neural Networks Council, 2010, 21 (10): 1624- 1642. doi: 10.1109/TNN.2010.2066988
14	蔡艳艳, 宋晓东. 针对非平衡数据分类的新型模糊SVM模型[J]. 西安电子科技大学学报, 2015, 42 (5): 134- 138, 174.
	CAI Y Y , SONG X D . A new fuzzy SVM model for imbalanced data classification[J]. Journal of Xidian University, 2015, 42 (5): 134- 138, 174.
15	王春玉, 苏宏业, 渠瑜, 等. 一种基于过抽样技术的非平衡数据集分类方法[J]. 计算机工程与应用, 2011, 47 (1): 139- 143.
	WANG C Y , SU H Y , QU Y , et al. Imbalanced data sets classification method based on over-sampling technique[J]. Computer Engineering and Applications, 2011, 47 (1): 139- 143.
16	张银峰, 郭华平, 职为梅, 等. 一种面向不平衡数据分类的组合剪枝方法[J]. 计算机工程, 2014, 40 (6): 157- 161, 165.
	ZHANG Y F , GUO H P , ZHI W M , et al. A classification method of combination pruning for unbalanced data[J]. Computer Engineering, 2014, 40 (6): 157- 161, 165.
17	VONG C M , IP W F , WONG P K , et al. Predicting minority class for suspended particulate matters level by extreme lear-ning machine[J]. Neurocomputing, 2014, 128 (27): 136- 144.
18	CHAWLA N V , BOWYER K W , HALL L O , et al. SMOTE: synthetic minority over-sampling technique[J]. Journal of Artificial Intelligence Research, 2011, 16 (1): 321- 357.
19	刘余霞, 刘三民, 刘涛, 等. 一种新的过采样算法DB_SMOTE[J]. 计算机工程与应用, 2014, 50 (6): 92- 95.
	LIU Y X , LIU S M , LIU T , et al. A new algorithm of over-sampling DB_SMOTE[J]. Computer Engineering and Application, 2014, 50 (6): 92- 95.
20	YANG Y , LIU F , JIN Z Y , et al. Aliasing artefact suppression in compressed sensing MRI for random phase-encode undersampling[J]. IEEE Trans.on Biomedical Engineering, 2015, 62 (9): 2215- 2223. doi: 10.1109/TBME.2015.2419372
21	JIA C Z , ZUO Y . S-SulfPred: a sensitive predictor to capture S-sulfenylation sites based on a resampling one-sided selection undersampling-synthetic minority oversampling technique[J]. Journal of Theoretical Biology, 2017, 422, 84- 89. doi: 10.1016/j.jtbi.2017.03.031
22	WILSON D L . Asymptotic properties of nearest neighbor rules using edited data[J]. IEEE Trans.on Systems Man & Cybernetics, 2007, SMC-2 (3): 408- 421.
23	赵自翔, 王广亮, 李晓东. 基于支持向量机的不平衡数据分类的改进欠采样方法[J]. 中山大学学报(自然科学版), 2012, 51 (6): 15- 21.
	ZHAO Z X , WANG G L , LI X D . An improved under-sampling method based on support vector machine for imbalanced data classification[J]. Acta Scientiarum Naturalium Universitatis Sunyatseni, 2012, 51 (6): 15- 21.
24	冯宏伟, 姚博, 高原, 等. 基于边界混合采样的非均衡数据处理算法[J]. 控制与决策, 2017, 32 (10): 1831- 1836.
	FENG H W , YAO B , GAO Y , et al. Imbalanced data processing algorithm based on boundary mixed sampling[J]. Control and Decision, 2017, 32 (10): 1831- 1836.
25	GUSTAVO E A P A B , RONALDO C P , MARIA C M . A study of the behavior of several methods for balancing machine learning training data[J]. ACM SIGKDD Explorations Newslet-ter, 2004, 6 (1): 20- 29. doi: 10.1145/1007730.1007735
26	谷琼, 袁磊, 宁彬, 等. 一种基于混合重取样策略的非平衡数据集分类算法[J]. 计算机工程与科学, 2012, 34 (10): 128- 134.
	GU Q , YUAN L , NING B , et al. A novel classification algorithm for imbalanced datasets based on hybrid resampling strategy[J]. Computer Engineering & Science, 2012, 34 (10): 128- 134.
27	陶新民, 郝思媛, 张冬雪, 等. 基于样本特性欠取样的不均衡支持向量机[J]. 控制与决策, 2013, 28 (7): 978- 984.
	TAO X M , HAO S Y , ZHANG D X , et al. Imbalanced support vector machines based on sample characteristics under-sampling[J]. Control and Decision, 2013, 28 (7): 978- 984.
28	SAVITHA R , SURESH S , SUNDARARAJAN N . Fast learning circular complex-valued extreme learning machine (CC-ELM) for real-valued classification problems[J]. Information Sciences, 2012, 187, 277- 290. doi: 10.1016/j.ins.2011.11.003
29	HUANG G B , ZHOU H M , DING X J , et al. Extreme learning machine for regression and multiclass classification[J]. IEEE Trans.on Systems, Man, and Cybernetics, 2012, 42 (2): 513- 529. doi: 10.1109/TSMCB.2011.2168604
30	薛丽香, 邱保志. 基于变异系数的边界点检测算法[J]. 模式识别与人工智能, 2009, 22 (5): 799- 802.
	XUE L X , QIU B Z . Boundary points detection algorithm based on coefficient of variation[J]. Pattern Recognition and Artificial Intelligence, 2009, 22 (5): 799- 802.
31	CRUZ R M O , SABOURIN R , CAVALCANTI G D C . META-DES Oracle: meta-learning and feature selection for dynamic ensemble selection[J]. Information Fusion, 2017, 38, 84- 103. doi: 10.1016/j.inffus.2017.02.010
32	PFAHRINGER B, BENSUSAN H, GIRAUD-CARRIER C G. Meta-learning by landmarking various learning algorithms[C]//Proc.of the 7th International Conference on Machine Learning, 2000: 743-750.

数据集	维数	少数类/多数类	不平衡比
Glass	9	39/175	1:4.487
Haberman	3	81/225	1:2.778
Ecoli	7	77/258	1:3.351
Breast	9	81/196	1:2.420

序号	V_{1_max}	V_{1_min}	V₂	V₃	V₄	V₅	V₆	V₇	V₈	属性
1	-7.730	-6.360	-6.923	-6.928	-6.281	-2.811	-2.981	-5.579	-0.140	正常
2	-7.794	-6.337	-6.953	-6.955	-6.297	-2.781	-2.969	-5.603	-0.134
⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮
188	-7.706	-6.344	-6.943	-6.945	-6.271	-2.812	-3.020	-5.613	-0.148
189	-7.760	-6.622	-7.106	-7.089	-6.533	-2.656	-2.456	-4.548	-0.133	故障
⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮
233	-7.792	-6.597	-7.078	-7.049	-6.503	-2.670	-2.544	-4.726	-0.113

数据集	算法	F-value		G-mean
数据集	算法	均值	标准差	均值	标准差
Glass	未处理	0.426	0.148	0.591	0.128
	SMOTE	0.479	0.153	0.573	0.126
	RU-SMOTE	0.458	0.176	0.565	0.172
	BMS	0.464	0.188	0.568	0.167
	RBCCS	0.667	0.110	0.653	0.106
Haberman	未处理	0.336	0.108	0.497	0.094
	SMOTE	0.445	0.160	0.517	0.168
	RU-SMOTE	0.432	0.187	0.536	0.184
	BMS	0.474	0.155	0.568	0.135
	RBCCS	0.639	0.081	0.633	0.069
Ecoli	未处理	0.344	0.102	0.492	0.084
	SMOTE	0.611	0.069	0.659	0.056
	RU-SMOTE	0.603	0.125	0.671	0.117
	BMS	0.430	0.118	0.546	0.097
	RBCCS	0.643	0.051	0.632	0.046
Breast	未处理	0.333	0.088	0.517	0.075
	SMOTE	0.961	0.028	0.943	0.027
	RU-SMOTE	0.925	0.072	0.931	0.069
	BMS	0.918	0.041	0.938	0.033
	RBCCS	0.950	0.022	0.950	0.021
Serial Regulator	未处理	0.498	0.166	0.693	0.129
	SMOTE	0.944	0.037	0.957	0.030
	RU-SMOTE	0.902	0.104	0.946	0.097
	BMS	0.931	0.039	0.952	0.028
	RBCCS	0.953	0.028	0.963	0.026

[1]	王峻洲, 王华伟, 侯召国. 基于相似体系的民机结构超手册维修案例分析[J]. 系统工程与电子技术, 2022, 44(9): 2978-2985.
[2]	童莹萍, 全英汇, 冯伟, 邢孟道. 基于空谱信息协同与Gram-Schmidt变换的多源遥感图像融合方法[J]. 系统工程与电子技术, 2022, 44(7): 2074-2083.
[3]	樊成, 王布宏, 田继伟. 基于多任务学习图卷积模型的航空网络节点分类[J]. 系统工程与电子技术, 2022, 44(7): 2341-2349.
[4]	曲志昱, 孙萌, 戴幻尧. 基于共形阵的角度和极化信息联合估计算法[J]. 系统工程与电子技术, 2022, 44(6): 1798-1804.
[5]	李海, 白锦, 孙研, 任嘉伟. 基于修正小波变换插值-TAN的雷达降水粒子分类[J]. 系统工程与电子技术, 2022, 44(5): 1527-1535.
[6]	李园, 史宪铭, 李亚娟, 赵美. 基于Adaboost的作战目标属性判定方法[J]. 系统工程与电子技术, 2022, 44(4): 1256-1262.
[7]	蔺瑞管, 王华伟, 车畅畅, 倪晓梅, 熊明兰. 基于LSTM分类器的航空发动机预测性维护模型[J]. 系统工程与电子技术, 2022, 44(3): 1052-1059.
[8]	赵筱彤, 周建江. 低截获MIMO雷达改进MUSIC算法[J]. 系统工程与电子技术, 2022, 44(2): 490-497.
[9]	吕勤哲, 全英汇, 沙明辉, 董淑仙, 邢孟道. 基于集成深度学习的有源干扰智能分类[J]. 系统工程与电子技术, 2022, 44(12): 3595-3602.
[10]	陈善学, 王欣欣. 基于空间预处理联合稀疏表示高光谱图像分类[J]. 系统工程与电子技术, 2021, 43(9): 2422-2429.
[11]	赖德迪, 罗智徽, 马应龙. 基于共现分析的分类器链标签序列优化方法[J]. 系统工程与电子技术, 2021, 43(9): 2526-2534.
[12]	丁嘉辉, 汤建龙, 于正洋. 轻量化的增量式集成学习算法设计[J]. 系统工程与电子技术, 2021, 43(4): 861-867.
[13]	陈唯实, 黄毅峰, 陈小龙, 卢贤锋, 张洁. 基于模型转换频率估计的低空目标分类[J]. 系统工程与电子技术, 2021, 43(4): 927-936.
[14]	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.
[15]	施伟, 黄红蓝, 冯旸赫, 刘忠. 面向多类别分类问题的子抽样主动学习方法[J]. 系统工程与电子技术, 2021, 43(3): 700-708.