基于改进多层核超限学习机的模拟电路故障诊断

doi:10.3969/j.issn.1000-1093.2021.02.013

摘要/Abstract

摘要： 为兼顾模拟电路多故障诊断的实用性和诊断精度，基于仿真诊断模型的测试性应用框架，结合深度学习与核方法的优势，提出一种多层单纯形优化核超限学习机（ML-SOKELM）方法。将有效初选后的数据集输入多层核超限学习机逐层提取故障特征并进行诊断；训练过程中，将各层核参数向量视为待优化变量，运用单纯形法对其进行联合优化。实验结果表明：与常见的深度学习方法相比，ML-SOKELM方法对主观经验依赖性更低，在训练时间大大缩短的同时，还能获得与之相当的准确率；与流行的核方法相比，ML-SOKELM方法在不同模糊度阈值下均能获得较高的诊断准确率。

关键词: 模拟电路, 核超限学习机, 深度学习, 故障诊断, 模糊组

Abstract: A multi-layer simplex optimized kernel extreme learning machine (ML-SOKELM) method is proposed based on the testability application framework of simulation-based diagnostic model, which combines the deep learning and kernel-based method. ML-SOKELM method is used to to improve the practicability and accuracy of multi-fault diagnosis of analog circuits. The multi-layer kernel extreme learning machine (ML-KELM) extracts the fault features layer by layer and gets the diagnosed results with original data after primary selection. During the training process, the proposed method is used to optimize the kernel parameters of all layers. The diagnosed results show that, compared with the common deep learning methods, ML-SOKELM methed is less dependent on subjective experience and achieves considerable accuracy while greatly shortening the training time. ML-SOKELM can achieve higher diagnostic accuracy under different ambiguity thresholds in comparison with popular kernel-based methods.

Key words: analogcircuit, kernelextremelearningmachine, deeplearning, faultdiagnosis, ambiguitygroup

中图分类号:

TM133

朱敏，许爱强，许晴，李睿峰. 基于改进多层核超限学习机的模拟电路故障诊断[J]. 兵工学报, 2021, 42(2): 356-369.

ZHU Min， XU Aiqiang， XU Qing， LI Ruifeng. Fault Diagnosis of Analog Circuits Based on Improved Multilayer Kernel Extreme Learning Machine[J]. Acta Armamentarii, 2021, 42(2): 356-369.

参考文献

［1］ ZHANG C L, HE Y G, YUAN L F, et al. Analog circuit incipi-ent fault diagnosis method using DBN based features extraction［J］. IEEE Access, 2018, 6(5): 23053-23064.
［2］ BINU D, KARIYAPPA B S. RideNN: a new rider optimization algorithm-based neural network for fault diagnosis in analog circuits［J］. IEEE Transactions on Instrumentation and Measurement, 2019, 68(1):2-26.
［3］ BINU D, KARIYAPPA B S. A survey on fault diagnosis of analog circuits: Taxonomy and state of the art［J］. AEU-International Journal of Electronics and Communications, 2017, 73:68-83.
［4］ TANG X F, XU A Q, LI R F, et al. Simulation-based diagnostic model for automatic testability analysis of analog circuits［J］. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2018,37(7): 1483-1493.
［5］ LUO H， LU W, WANG Y R, et al. A new test point selection method for analog continuous parameter fault［J］. Journal of Electronic Testing: Theory & Applications, 2017, 33(3):339-352.
［6］ TANG X F, XU A Q, NIU S C. KKCV-GA-based method for optimal analog test point selection［J］. IEEE Transactions on Instrumentation and Measurement, 2017, 66(1):24-32.
［7］ KUMAR A， SINGH A P. Fuzzy classifier for fault diagnosis in analog electronic circuits［J］. ISA Transactions, 2013, 52(6):816-824.
［8］ LITOVSKI V, ANDREJEVIAC＇U2 M, ZWOLINSKI M. Analogue electronic circuit diagnosis based on ANNs［J］. Microelectronics Reliability, 2006, 46(8):1382-1391.
［9］袁莉芬, 孙业胜, 何怡刚, 等. 基于小波包优选的模拟电路故障特征提取方法［J］. 电工技术学报, 2018, 33(1):158-165.
YUAN L F, SUN Y S, HE Y G, et al. Fault feature extraction method for analog circuit based on preferred wavelet packet ［J］. Transactions of China Electrotechnical Society, 2018, 33(1): 158- 165. (in Chinese)
［10］ LIU Z B, LIU T M, HAN J W, et al. Signal model-based fault coding for diagnostics and prognostics of analog electronic circuits［J］. IEEE Transactions on Industrial Electronics, 2017, 64(1): 605-614.
［11］ XIE T， HE Y G. Fault diagnosis of analog circuit based on high-order cumulants and information fusion［J］. Journal of Electronic Testing: Theory & Applications, 2014, 30(5):505-514.
［12］ YANG H H, MENG C, WANG C. Data-driven feature extraction for analog circuit fault diagnosis using 1-D convolutional neural network ［J］. IEEE Access, 2020, 8(1): 18305-18315.
［13］ LIU Z B, JIA Z, VONG C M, et al. Capturing high-discriminative fault features for electronics-rich analog system via deep learning［J］. IEEE Transactions on Industrial Informatics, 2017, 13(3):1213-1226.
［14］ XU G W, LIU M, JIANG Z F, et al. Online fault diagnosis method based on transfer convolutional neural networks［J］. IEEE Transactions on Instrumentation and Measurement, 2020, 69(2):509-520.
［15］廖剑, 史贤俊, 周绍磊, 等. 基于局部图嵌入加权罚SVM的模拟电路故障诊断方法［J］. 电工技术学报, 2016, 31(4): 28-35.
LIAO J, SHI X J, ZHOU S L, et al. Analog circuit fault diagnosis based on local graph embedding weighted-penalty SVM［J］. Transactions of China Electrotechnical Society, 2016, 31(4): 28- 35. (in Chinese)
［16］张伟, 刘星, 许爱强, 等. lp范数约束的模拟电路3层多核故障诊断模型［J］. 兵工学报, 2018, 39(7):1352-1363.
ZHANG W, LIU X, XU A Q, et al. Three-layer multiple kernel fault diagnosis model with lp-norm constraint for analog circuit［J］. Acta Armamentarii, 2018, 39(7):1352-1363. (in Chinese)
［17］高明哲,许爱强,唐小峰, 等.基于多核多分类相关向量机的模拟电路故障诊断方法［J］.自动化学报,2019,45(2):434-444.
GAO M Z, XU A Q, TANG X F, et al. Analog circuit diagnostic method based on multi-kernel learning multiclass relevance vector machine［J］. Acta Automatica Sinica, 2019，45(2):434-444. (in Chinese)
［18］ LIYANAARACHCHI L C K, ZHOU H M, HUANG G B. Representational learning with ELMs for big data［J］. IEEE Intelligent Systems,2013,28(6): 31-34.
［19］ KASUN L C， YANG Y， HUANG G B， et al. Dimension reduction with extreme learning machine［J］. IEEE Transactions on Image Processing, 2016, 25(8):3906-3918.
［20］ TANG J X, DENG C W, HUANG G B. Extreme learning machine for multilayer perception［J］. IEEE Transactions on Neural Networks and Learning Systems, 2016,27(4):809-821.
［21］逄珊, 杨欣毅, 张勇, 等. 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 S, YANG X Y, ZHANG Y， et al. Application of deep kernel extreme learning machinein aero engine components fault diagnosis［J］. Journal of Propulsion Technology, 2017,38(11): 2613-2621. (in Chinese)
［22］ TANG X F, XU A Q. Practical analog circuit diagnosis based on fault features with minimum ambiguities［J］. Journal of Electro-nic Testing, 2016, 32(1):83-95.
［23］ CHEN X, WANG W, CAO W, et al. Gaussian-kernel-based adaptive critic design using two-phase value iteration［J］. Information Sciences, 2019, 482(6): 139-155.
［24］ LAGARIAS J C, REEDS J A, WRIGHT M H, et al. Convergence properties of the Nelder-Mead simplex method in low dimensions［J］. SIAM Journal on Optimization: A Publication of the Society for Industrial & Applied Mathematics, 2006, 9(1):112-147.
［25］ IOSIFIDIS A, TEFAS A, PITAS I. Graph embedded extreme learning machine［J］. IEEE Transactions on Cybernetics, 2016,46(1):311-324.
［26］ PSORAKIS I， DAMOULAS T, GIROLAMI M A. Multiclass relevance vector machines: sparsity and accuracy［J］. IEEE Tran-sactions on Neural Networks, 2010, 21(10):1588-1598.

[1]	李良福，陈卫东，高强，许开銮，刘轩，何曦，钱钧. 基于深度学习的光电系统智能目标识别[J]. 兵工学报, 2022, 43(S1): 162-168.
[2]	李良钰，苏铁熊，马富康，吴小军，徐春龙. 基于集合经验模态分解-支持向量机的高压共轨系统故障诊断方法[J]. 兵工学报, 2022, 43(5): 992-1001.
[3]	李锦青，刘泽飞，满振龙. 基于生成对抗网络的密钥生成方法及其在微光图像加密中的应用[J]. 兵工学报, 2022, 43(2): 337-344.
[4]	普运伟，刘涛涛，郭江，吴海潇. 基于卷积神经网络和模糊函数主脊坐标变换的雷达辐射源信号识别[J]. 兵工学报, 2021, 42(8): 1680-1689.
[5]	李志农, 胡志峰, 毛清华, 张旭辉, 陶俊勇. 非线性调频模态分解-同步提取变换方法及其在滚动轴承故障诊断中的应用[J]. 兵工学报, 2021, 42(6): 1324-1330.
[6]	张凤, 张超, 杨华民, 王发斌. 作战辅助系统中基于深度学习的投影图像颜色补偿方法[J]. 兵工学报, 2021, 42(11): 2418-2423.
[7]	薛灵芝, 曾向阳, 杨爽. 基于生成对抗网络的水声目标识别算法[J]. 兵工学报, 2021, 42(11): 2444-2452.
[8]	余跃，王宏伦. 基于深度学习的高超声速飞行器再入预测校正容错制导[J]. 兵工学报, 2020, 41(4): 656-669.
[9]	闫孟达，杨任农，左家亮，胡东愿，岳龙飞，赵雨. 基于深度学习的空空导弹多类攻击区实时解算[J]. 兵工学报, 2020, 41(12): 2466-2477.
[10]	梁杰，任君，李磊，齐航，周红丽. 基于典型几何形状精确回归的机场跑道检测方法[J]. 兵工学报, 2020, 41(10): 2045-2054.
[11]	余永维，彭西，杜柳青，陈天皓. 基于深度学习框架的装配机器人零件实时检测方法[J]. 兵工学报, 2020, 41(10): 2122-2130.
[12]	李佳蔚，崔涛，刘宇航，师帅楠，孙强. 基于排气温度动态模型的在线观测器研究[J]. 兵工学报, 2019, 40(8): 1562-1571.
[13]	闫理跃, 王厚军, 刘震. 一种结合噪声辅助技术和现场数据的模拟电路实时可靠度预测方法[J]. 兵工学报, 2019, 40(2): 326-333.
[14]	郭伟超，赵怀山，李成，李言，汤奥斐. 基于小波包能量谱与主成分分析的轴承故障特征增强诊断方法[J]. 兵工学报, 2019, 40(11): 2370-2377.
[15]	翟禹尧，史贤俊，吕佳朋. 基于广义随机Petri网的导弹系统测试性建模与指标评估方法研究[J]. 兵工学报, 2019, 40(10): 2070-2079.