﻿ 基于粗集理论的舰船机电设备故障诊断研究
 舰船科学技术  2022, Vol. 44 Issue (16): 114-117    DOI: 10.3404/j.issn.1672-7649.2022.16.023 PDF

Research on fault diagnosis of ship electromechanical equipment based on rough set theory
XU Xin
Department of Mechanical Engineering, Taiyuan Institute of Technology, Taiyuan 030008, China
Abstract: Ship electromechanical equipment is an important equipment for ship operation. When the electromechanical equipment fails, it directly threatens the safety of ship navigation. The ship's electromechanical equipment is complex in composition, and there is a nonlinear relationship between each equipment. A large amount of data information will be generated in a short period of time. It is necessary to use an algorithm with good timeliness and high accuracy to assist in the completion of fault diagnosis. Based on this, based on the overview of rough set theory and support vector machine fault diagnosis method, this paper proposes a fault diagnosis method for ship electromechanical equipment based on rough set theory. In the diagnosis, the rough set theory and the improved SVM model are combined. The experiments show that The diagnostic accuracy of this electromechanical equipment diagnostic model is over 96%.
Key words: rough set theory     ship     electromechanical equipment     fault diagnosis
0 引　言

1 粗集理论概述 1.1 基础理论

 $P \in Q,R \in Q,R \in P \text{，}$

1.2 属性约简方法

 $a = \frac{1}{{k + 2}}\left[ {{a_c}(L) + {\gamma _c}(L) + \sum\limits_{j + 1}^k {{a_c}({Y_j})} } \right] \text{，}$

2 支持向量机故障诊断方法 2.1 SVM模型

2.2 核函数运用

2.3 松弛变量确定

3 基于粗集理论的舰船机电设备故障诊断 3.1 粗集属性约简

 图 1 基于粗集理论的机电设备故障属性约简流程图 Fig. 1 Flow chart of fault attribute reduction of electromechanical equipment based on rough set theory

3.2 故障诊断分析

 图 2 基于粗集理论改进后SVM的舰船机电设备故障诊断故障特征值分布图 Fig. 2 Distribution of fault eigenvalues of ship electromechanical equipment fault diagnosis based on improved SVM based on rough set theory

 图 3 基于改进后SVM二类分类器拓扑结构的节点位置示意图 Fig. 3 Schematic diagram of node location based on the topology of improved SVM class II classifier

 图 4 基于改进后SVM二类分类器的最优参数分析图 Fig. 4 Analysis diagram of optimal parameters based on Improved SVM class II classifier

 图 5 舰船主机设备故障诊断测试集的实际分类示意图 Fig. 5 Schematic diagram of the actual classification of the fault diagnosis test set of the ship's main equipment
3.3 优化SVM拓扑结构

4 结　语

 [1] 李锐君, 胡代弟, 侯维岩. 基于无线通信技术的自动化分拣机械臂故障检测系统[J]. 制造业自动化, 2022, 44(4): 158-162. LI Rui-jun, HU Dai-di, HOU Wei-yan. Automatic sorting robot arm fault detection system based on wireless communication technology[J]. Manufacturing Automation, 2022, 44(4): 158-162. DOI:10.3969/j.issn.1009-0134.2022.04.036 [2] 张富魁. 基于小波分析的煤矿机械设备运行故障检测方法[J]. 内蒙古煤炭经济, 2022(5): 59-61. ZHANG Fu-kui. Fault detection method for coal mine machinery and equipment operation based on wavelet analysis[J]. Inner Mongolia Coal Economy, 2022(5): 59-61. DOI:10.13487/j.cnki.imce.021855 [3] 田睿. 农机异步电机故障检测系统研究与设计[J]. 南方农机, 2022, 53(10): 61-63+72. TIAN Rui. Research and design of asynchronous motor fault detection system for agricultural machinery[J]. Nanfang Agricultural Machinery, 2022, 53(10): 61-63+72. DOI:10.3969/j.issn.1672-3872.2022.10.017 [4] 李佩青, 李晓波, 沈腾. 船用柴油机颗粒捕集器主动再生技术研究[J]. 内燃机与配件, 2022(11): 9-12. LI Pei-qing, LI Xiao-bo, SHEN Teng. Research on active regeneration technology of marine diesel engine particulate filter[J]. Internal Combustion Engine and Parts, 2022(11): 9-12. DOI:10.19475/j.cnki.issn1674-957x.2022.11.005