作者：李楠1,2, 邓威3, 王晨4, 吴光辉5
作者单位：1. 南京航空航天大学航空学院，江苏 南京 210016;
2. 中国商飞民用飞机试飞中心试飞运行部，上海 201323;
3. 国防科技大学卫星导航研发中心，湖南 长沙 410073;
4. 中国商飞北京民用飞机技术研究中心，北京 102211;
5. 太原理工大学，山西 太原 030024
关键词：航空电子;模拟电路;故障诊断;概率神经网络
摘要：
模拟电路已广泛应用于航空电子系统，模拟电路的失效会影响系统的功能，引起系统故障，甚至引发灾难性的安全事故。为快速准确地实现模拟电路的故障诊断，该文引入概率神经网络方法，并针对传统概率神经网络方法中的诊断准确性、诊断效率问题，提出基于K-means与概率神经网络的模拟电路故障诊断方法，定义聚类有效性指标，采用K-means聚类分析与有效性指标分析相结合的方式，选取聚类中心作为模式层神经元训练概率神经网络模型，从而降低模型的复杂程度，大大减少故障诊断时间。最后，以有源滤波电路为对象，通过与传统概率神经网络方法以及随机概率神经网络方法的对比分析，验证该文方法在故障诊断准确性以及故障诊断效率上的优越性能。
Analog circuit fault diagnosis method based on K-means and probabilistic neural network
LI Nan1,2, DENG Wei3, WANG Chen4, WU Guanghui5
1. College of Aerospace Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China;
2. Flight Test Operation Department, Flight Test Center of COMAC, Shanghai 201323, China;
3. National University of Defense Technology Satellite Navigation Research and Development Center, Changsha 410073, China;
4. COMAC Beijing Aircraft Technology Research Institute, Beijing 102211, China;
5. Taiyuan University of Technology, Taiyuan 030024, China
Abstract: Analog circuit has been widely used in avionics system. The failure of analog circuit will affect the function of the system, cause system failure and even catastrophic accident. This paper proposes a combined method based on K-means and probabilistic neural network for analog circuit fault diagnosis. In view of the low diagnostic accuracy and efficiency of traditional probabilistic neural network, K-means clustering analysis is adopted to select clustering centers as pattern neurons. Thus, the complexity of the diagnostic model and the fault diagnosis time are reduced. Finally, the proposed method is compared with traditional probabilistic neural network and random probabilistic neural network, the experimental results of an active filter circuit verify the superior performance of the method in fault diagnosis accuracy and fault diagnosis efficiency.
Keywords: avionics;analog circuit;fault diagnosis;probabilistic neural network
2021, 47(3):98-103,109 收稿日期: 2020-06-12;收到修改稿日期: 2020-08-12
基金项目:
作者简介: 李楠(1983-),女,吉林四平市人,高级工程师,硕士,研究方向为飞机发动机故障诊断
参考文献
[1] ZHANG C, HE Y, YUAN L, et al. A novel approach for diagnosis of analog circuit fault by using GMKL-SVM and PSO[J]. Journal of Electronic Testing, 2016, 32(5): 531-540
[2] 蒋海峰, 张曼, 赵斌炎, 等. 基于改进Hilbert-Huang变换的电网故障诊断[J]. 电工技术学报, 2019, 34(S1): 336-342
[3] 杨志淳, 沈煜, 杨帆, 等. 考虑多元因素态势演变的配电变压器迁移学习故障诊断模型[J]. 电工技术学报, 2019, 34(7): 1505-1515
[4] 陈勇, 梁洪, 王成栋, 等. 基于改进小波包变换和信号融合的永磁同步电机匝间短路故障检测[J]. 电工技术学报, 2020, 34(S1): 228-234
[5] 祝文姬, 何怡刚. 容差模拟电路软故障诊断的神经网络方法[J]. 电工技术学报, 2009, 24(11): 184-191
[6] 钟建林, 何友, 任献彬. 基于波形相似度的容差模拟电路软故障诊断[J]. 电工技术学报, 2012, 27(8): 222-229,
[7] 董海迪, 刘刚, 何兵, 等. 容差模拟电路参数故障诊断[J]. 中国测试, 2017, 43(9): 128-133
[8] 肖迎群, 冯良贵, 何怡刚. 基于小波分形和核判别分析的模拟电路故障诊断[J]. 电工技术学报, 2012, 27(8): 230-238
[9] 王淑娟, 陈博, 赵国良. 基于小波包变换预处理的模拟电路故障诊断方法[J]. 电工技术学报, 2003(4): 118-122
[10] 唐志军, 何怡刚, 詹杰, 等. 基于符号分析的模拟电路故障诊断方法及实现[J]. 电工技术学报, 2008(1): 24-29
[11] SHOKROLAHI S M, KAZEMPOUR A T N. A novel approach for fault detection of analog circuit by using improved EEMD[J]. Analog Integrated Circuits and Signal Processing, 2019, 98(3): 527-534, Mar
[12] XIAO Y, HE Y. A novel approach for analog fault diagnosis based on neural networks and improved kernel PCA[J]. Neurocomputing, 2011, 74(7): 1102-1115
[13] 朱敏, 许晴, 唐小峰, 等. 基于KDE-FPBT的模拟电路测试性分析与故障诊断方法研究[J]. 中国测试, 2018, 44(2): 11-15
[14] 董海迪, 刘刚, 何兵, 等. 基于系统搜索的容差模拟电路多参数故障诊断[J]. 中国测试, 2019, 45(7): 134-139
[15] 董海迪, 王俊提, 刘刚, 等. 基于迭代求解的容差模拟电路多参数故障诊断[J]. 中国测试, 2018, 44(4): 31-36
[16] 简春菲, 王晓峰, 王司. 基于ALO-SVM的模拟电路故障诊断研究[J]. 自动化技术与应用, 2019, 38(12): 113-118
[17] ZHANG C, HE Y, DENG F, et al. A novel approach for analog circuit fault prognostics based on improved RVM[J]. Journal of Electronic Testing, 2014, 30(3): 343-356
[18] SHEIKHAN M, SHA A A. PSO-optimized modular neural network trained by OWO-HWO algorithm for fault location in analog circuits[J]. Neural Computing & Applications, 2013, 23(2): 519-530
[19] 何怡刚, 祝文姬, 周炎涛, 等. 基于粒子群算法的模拟电路故障诊断方法[J]. 电工技术学报, 2010, 25(6): 163-171
[20] LUO H, WANG Y, LIN H, et al. Module level fault diagnosis for analog circuits based on system identification and genetic algorithm[J]. Measurement, 2012, 45(4): 769-777
[21] YANG C, ZHEN L, HU C. Fault diagnosis of analog filter circuit based on genetic algorithm[J]. IEEE Access, 2020, PP(99): 1-1
[22] AMINIAN M, AMINIAN F. A modular fault-diagnostic system for analog electronic circuits using neural networks with wavelet transform as a preprocessor[J]. IEEE Transactions on Instrumentation & Measurement, 2007, 56(5): 1546-1554
[23] AMINIAN F, AMINIAN M, COLLIONS H W. Analog fault diagnosis of actual circuits using neural networks[J]. IEEE Transactions on Instrumentation & Measurement, 2002, 51(3): 544-550
[24] 韩海涛, 马红光, 曹建福, 等. 基于非线性频谱特征及核主元分析的模拟电路故障诊断方法[J]. 电工技术学报, 2012, 27(8): 248-254
[25] BILSKI P. Hierarchical diagnostics of analog systems based on the ambiguity groups detection[J]. Measurement, 2018, 119: 1-10
