Infrared Image-based ISSA-BP Neural Network for Airborne Circuit Board Chip Fault Diagnosis
-
摘要: 针对传统红外图像的机载电路板芯片故障诊断法诊断率低且无法诊断动态故障的问题,本文提出了一种基于红外温度数据的改进麻雀搜索算法优化BP神经网络(Improved sparrow search algorithm-Back propagation neural networks, ISSA-BPNN)机载电路板芯片故障诊断方法。首先,提取红外热像仪采集的电路板芯片温度数据,建立电路板芯片升温过程中静态、动态、统计特征的特征模型;然后,利用Sine混沌映射初始化麻雀种群分布,利用Levy飞行策略改进发现者种群位置更新公式,将改进后的麻雀搜索算法优化BP神经网络的权值参数;最后,将温度特征模型输入到ISSA-BP神经网络进行训练和测试,从而完成电路板芯片故障诊断。实验采用航电系统电源电路板进行可靠性分析,实验结果表明,该方法在电路板不同工况下综合故障诊断率达到97.84%。Abstract: This study proposes an improved sparrow search algorithm based on an infrared temperature-data-optimized back propagation neural network (ISSA-BPNN) for an airborne circuit board chip fault diagnosis method that cannot diagnose dynamic failures. First, an infrared thermal imaging camera collected circuit board chip temperature data to establish a feature model of static, dynamic, and statistical characteristics of the circuit board chip warming process. We used sine chaos mapping to initialize the sparrow population distribution, the levy flight improvement finder sparrow location update, and an improved sparrow search algorithm to optimize the weight parameters of the BP neural network. Finally, the temperature feature model was input to the ISSA-BP neural network for training and testing to complete the circuit board chip fault diagnosis. The experiments used an avionics system power supply circuit board for reliability analysis, and the results revealed that the method achieved a comprehensive fault diagnosis rate of 97.84% under different circuit board operating conditions.
-
表 1 故障模式总数据集构成
Table 1. Failure mode total data set composition
Mode Chip Number Fault Mode Samples F1 None Normal 90 F2~F3 U1 Open 180 F4~F7 U2 Open 360 F8~F9 U3 Open 180 F10~F12 U1 Virtual Soldering 270 F13~F15 U2 Virtual Soldering 270 F16~F18 U3 Virtual Soldering 270 F19~F21 U1 Short 270 F22~F24 U2 Short 270 F25~F27 U3 Short 270 表 2 低功率空载故障诊断率对比
Table 2. Low power empty load fault diagnosis rate comparison
Accuracy/% F1 F2~F9 F10~F18 F19~F27 BP 100 84.38 88.89 86.11 SSA-BP 100 90.63 91.67 88.89 ISSA-BP 100 96.88 97.22 97.22 表 3 低功率1020 Ω故障诊断率对比
Table 3. Low power 1020 Ω fault diagnosis rate comparison
Accuracy/% F1 F2~F9 F10~F18 F19~F27 BP 75 90.63 77.78 83.33 SSA-BP 100 94.44 83.33 91.67 ISSA-BP 100 100 97.22 100 表 4 低功率330 kΩ故障诊断率对比
Table 4. Low-power 330 kΩ fault diagnosis rate comparison
Accuracy/% F1 F2~F9 F10~F18 F19~F27 BP 75 93.75 83.33 86.11 SSA-BP 100 96.88 88.89 88.89 ISSA-BP 100 100 94.44 100 表 5 高功率空载故障诊断率对比
Table 5. High-power empty load fault diagnosis rate comparison
Accuracy/% F1 F2~F9 F10~F18 F19~F27 BP 100 68.75 58.33 61.11 SSA-BP 100 81.25 69.44 83.33 ISSA-BP 100 96.88 97.22 97.22 表 6 高功率1020 Ω故障诊断率对比
Table 6. High-power 1020 Ω fault diagnosis rate comparison
Accuracy/% F1 F2~F9 F10~F18 F19~F27 BP 50 65.63 69.44 83.33 SSA-BP 75 83.33 80.56 88.89 ISSA-BP 100 96.88 97.22 97.22 表 7 高功率330 kΩ故障诊断率对比
Table 7. High-power 330 kΩ fault diagnosis rate comparison
Accuracy/% F1 F2~F9 F10~F18 F19~F27 BP 75 87.5 80.56 91.67 SSA-BP 100 90.63 86.11 94.44 ISSA-BP 100 96.88 97.22 100 表 8 多工况下3种故障分别诊断率
Table 8. Diagnosis rate of three faults separately under multiple working conditions
Accuracy
/%Open
(F2~F9)Virtual soldering
(F10~F18)Short
(F19~F27)BP 81.77 76.39 81.94 SSA-BP 88.54 83.33 89.35 ISSA-BP 97.92 96.76 98.61 表 9 多工况下故障诊断率对比
Table 9. Comparison of fault diagnosis rate under multiple working conditions
Mode Results Combined accuracy BP 514/648 79.32% SSA-BP 566/648 87.35% ISSA-BP 634/648 97.84% -
[1] SHI L, ZHOU Z, JIA H, et al. Fault diagnosis of functional circuit in avionics system based on BPNN[C]//Prognostics and System Health Management Conference, IEEE, 2019: 1-5. [2] DONG Z, CHEN L. Image registration in PCB fault detection based on infrared thermal imaging[C]//Chinese Control Conference (CCC). IEEE, 2019: 4819-4823. [3] Sarawade A A, Charniya N N. Detection of faulty integrated circuits in PCB with thermal image processing[C]//International Conference on Nascent Technologies in Engineering (ICNTE). IEEE, 2019: 1-6. [4] 孙东旭, 贾世伟, 孟玉慈, 等. 综合模块化航电系统FC网络的机内测试设计[J]. 航空计算技术, 2016, 46(6): 108-112. https://www.cnki.com.cn/Article/CJFDTOTAL-HKJJ201606028.htmSUN Dongxu, JIA Shiwei, MENG Yuci, et al. Built-in-test design for integrated module avionics fibre channel networks[J]. Aeronautical Computing Technique, 2016, 46(6): 108-112. https://www.cnki.com.cn/Article/CJFDTOTAL-HKJJ201606028.htm [5] ZHENG T, WANG Z, TAN C, et al. Research on fault prediction and diagnosis method of PCB circuit[C]//International Conference on Artificial Intelligence and Computer Engineering (ICAICE). IEEE, 2020: 387-390. [6] 徐奇伟, 黄宏, 张雪锋, 等. 基于改进区域全卷积网络的高压引线接头红外图像特征分析的在线故障诊断方法[J]. 电工技术学报, 2021, 36(7): 1380-1388. https://www.cnki.com.cn/Article/CJFDTOTAL-DGJS202107008.htmXU Qiwei, HUANG Hong, ZHANG Xuefeng, et al. Online fault diagnosis method for infrared image feature analysis of high-voltage lead connectors based on improved R-FCN[J]. Transactions of China Electrotechnical Society, 2021, 36(7): 1380-1388. https://www.cnki.com.cn/Article/CJFDTOTAL-DGJS202107008.htm [7] 张博. 基于红外图像电路板故障检测技术研究[D]. 沈阳: 沈阳航空航天大学, 2018.ZHANG Bo. Research on Circuit Board Fault Detection Technology Based on Infrared Image[D]. Shenyang: Shenyang Aerospace University, 2018. [8] Salvi S S, Jain A. Detection of unusual thermal activities in a semiconductor chip using backside infrared thermal imaging[J]. Journal of Electronic Packaging, 2021, 143(2): 020901. doi: 10.1115/1.4049291 [9] Al Obaidy F, Yazdani F, Mohammadi F A. Intelligent testing for Arduino UNO based on thermal image[J]. Computers & Electrical Engineering, 2017, 58: 88-100. [10] 王力, 李硕, 武会杰, 等. 基于红外技术的机载电路板故障诊断研究[J]. 计算机仿真, 2017, 34(7): 89-93. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJZ201707019.htmWANG Li, LI Shuo, WU Huijie, et al. Research of aviation circuit board fault diagnosis and prediction based on infrared technology[J]. Computer Simulation, 2017, 34(7): 89-93. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJZ201707019.htm [11] 郝建新, 贾春宇. 基于红外热图的机载电路板故障模式诊断研究[J]. 红外技术, 2019, 41(3): 273-278. http://hwjs.nvir.cn/article/id/hwjs201903013HAO Jianxin, JIA Chunyu. Research on fault mode diagnosis of airborne circuit board based on infrared images[J]. Infrared Technology, 2019, 41(3): 273-278. http://hwjs.nvir.cn/article/id/hwjs201903013 [12] 王力, 刘子奇. WPA-IGA-BP神经网络的模拟电路故障诊断[J]. 系统工程与电子技术, 2021, 43(4): 1133-1143. https://www.cnki.com.cn/Article/CJFDTOTAL-XTYD202104034.htmWANG Li, LIU Ziqi. Fault diagnosis of analog circuit for WPA-IGA-BP neural network[J]. System Engineering and Electronics, 2021, 43(4): 1133-1143. https://www.cnki.com.cn/Article/CJFDTOTAL-XTYD202104034.htm [13] GUO Y, RAN C, JI X, et al. Fault diagnosis in analog circuits based on combined-optimization BP neural networks[J]. Journal of Northwestern Polytechnical University, 2013, 1: 44-48. [14] XUE J, SHEN B. A novel swarm intelligence optimization approach: sparrow search algorithm[J]. Systems Science & Control Engineering An Open Access Journal, 2020, 8(1): 22-34. [15] Reynolds Andy. Liberating Lévy walk research from the shackles of optimal foraging[J]. Physics of Life Reviews, 2015, 14: 59-83.