﻿ 基于人工鱼群算法的船舶电子设备故障智能诊断方法
 舰船科学技术  2023, Vol. 45 Issue (24): 188-191    DOI: 10.3404/j.issn.1672-7649.2023.24.035 PDF

Intelligent fault diagnosis method of ship electronic equipment based on artificial fish swarm algorithm
RUAN Jia
Wuhan Technical College of Communications, Wuhan 430065, China
Abstract: Research on the intelligent diagnosis method for ship electronic equipment faults using artificial fish swarm algorithm, to achieve efficient maintenance of electronic equipment and ensure safe navigation of ships. The discrete wavelet transform method is used to decompose the operating signal samples of electronic equipment. After extracting the fault feature parameters of ship electronic equipment by calculating the wavelet energy values at different scales, it is used as input for the fault diagnosis model based on RBF neural network. The artificial fish swarm algorithm is used to optimize the weight and threshold parameters of the fault diagnosis model, and ultimately output the probability of different types of faults to achieve electronic equipment fault diagnosis. The experimental results show that there are significant differences in the time-domain waveforms of electronic equipment operating signals under normal and different fault states. The research method can extract fault feature parameters and complete the identification of fault types. After 30 iterations, the MSE index can be reduced to the lowest, only 10−4.
Key words: artificial fish school algorithm     electronic equipment     fault diagnosis     discrete wavelet transform     RBF neural network     characteristic parameter
0 引　言

1 船舶电子设备故障智能诊断 1.1 船舶电子设备故障特征参数提取

 ${a_i}\left( k \right) = \left\langle {X,{\phi _{ij}}\left( t \right)} \right\rangle ，$ (1)
 ${d_i}\left( k \right) = \left\langle {X,{\varPsi _{ij}}\left( t \right)} \right\rangle。$ (2)

2 实验结果分析

 图 2 电台设备正常、故障状态下的时域波形 Fig. 2 Time domain waveforms of radio equipment under normal and fault conditions

 图 3 故障诊断模型性能评价 Fig. 3 Performance evaluation of fault diagnosis model
3 结　语

1）电子设备处于不同运行状态下，其信号时域波形曲线差异很大。

2）本文方法可实现船舶电子设备故障诊断，MSE指标仅为10−4

 [1] 吴旭升, 杨刚, 孙盼, 等. 舰船电力电子设备测试性设计现状及技术分析[J]. 海军工程大学学报, 2021, 33(2): 14-18. WU Xusheng, YANG Gang, SUN Pan, et al. Analysis of status and technology of testability design for ship power and electronic equipment[J]. Journal of Naval University of Engineering, 2021, 33(2): 14-18. DOI:10.7495/j.issn.1009-3486.2021.02.003 [2] 单显明, 李长伟, 张忠传. 基于神经网络的复杂电子设备故障诊断系统的设计[J]. 电子测量技术, 2022, 45(11): 52-56. SHAN Xianming, LI Changwei, ZHANG Zhongchuan. Design of fault diagnosis system of complex electronic equipment based on neural network[J]. Electronic Measurement Technology, 2022, 45(11): 52-56. [3] 盛沛, 许爱强, 单鑫, 等. 基于边际能量谱的电子设备性能退化型故障诊断方法研究[J]. 国外电子测量技术, 2020, 39(2): 118-122. SHENG Pei, XU Aiqiang, SHAN Xin, et al. Feature vector extraction in performance degradation analog circuit based on marginal spectrum[J]. Foreign Electronic Measurement Technology, 2020, 39(2): 118-122. DOI:10.19652/j.cnki.femt.1901836 [4] 罗德强, 王汉军, 牛春辉. 基于神经网络算法的某型装备专家诊断系统的研究[J]. 舰船电子工程, 2020, 40(12): 127-129. LUO Deqiang, WANG Hanjun, NIU Chunhui. Study on expert diagnosis system of certain equipment based on neural network algorithms[J]. Ship Electronic Engineering, 2020, 40(12): 127-129. DOI:10.3969/j.issn.1672-9730.2020.12.029 [5] 杨英楠, 庄新伟, 韩金良. 舰船电子设备节点模块级硬件故障诊断方法研究[J]. 舰船电子工程, 2021, 41(3): 150-152+161. YANG Yingnan, ZHUANG Xinwei, HAN Jinliang. Research on node module level hardware fault diagnosis method of ship electronic equipment[J]. Ship Electronic Engineering, 2021, 41(3): 150-152+161. [6] 武剑, 薛玉石, 山春凤, 等. 基于改进人工鱼群优化算法的配网故障定位研究[J]. 电力科学与技术学报, 2023, 38(2): 40-47. WU Jian, XUE Yushi, SHAN Chunfeng, et al. A location approach for distribution network faults based on the enhanced artificial fish swarm optimization[J]. Journal of Electric Power Science and Technology, 2023, 38(2): 40-47.