基于EMD-SVM的航空故障电弧检测
[Abstract]:With the development of aviation industry in China, the safe operation of aeronautical electrical system has attracted more and more attention. However, the working environment of aeronautical cable is special, and it needs to work in the environment of high temperature, high radiation and high vibration for a long time. These factors will accelerate the aging of aeronautical cable and induce fault arc. Aeronautical fault arc has high energy, and it is easy to cause fire, which is a serious threat to aviation safety. Therefore, the research of aeronautical fault arc detection technology is an important subject to ensure aviation safety. In this paper, the generation mechanism of aeronautical fault arc is summarized firstly, and the Cassie dynamic arc model is used to simulate the aeronautical series arc in Simulink environment. Based on the test method of fault arc in UL1699 standard, the test scheme is designed and the aeronautical arc test platform is built. The test platform is used to collect a large number of test data of aeronautical fault arc, and the fault arc test data of resistive load, inductive load and nonlinear load under the condition of 400Hz are obtained, and the aeronautical arc test database is established. It lays a foundation for the research of fault detection algorithm of aeronautical arc. In order to extract the fault characteristics of aeronautical arc, after analyzing the characteristics of arc current in frequency domain, the empirical mode decomposition (EMD) method is introduced to stabilize the current waveform of aeronautical arc. Then the third order autoregressive (AR) model is established for the IMF component of intrinsic mode function obtained by EMD, and the parameters of AR model are estimated by Burg algorithm. The model parameters can reflect the fault characteristics effectively and can be used as the eigenvector of aeronautical fault arc. Finally, the least square support vector machine (LS-SVM) is applied to the recognition of aeronautical fault arc, and the LS-SVM learning machine is constructed. The processed data are divided into training set and test set, and the LS-SVM classifier is trained and tested to identify the aero-arc fault of linear load, nonlinear load and unknown load. The results show that the algorithm can effectively identify the arc faults of nonlinear load and unknown load, and can provide a reference for the detection of aeronautical series fault arc.
【学位授予单位】:大连理工大学
【学位级别】:硕士
【学位授予年份】:2016
【分类号】:V267
【参考文献】
中国期刊全文数据库 前10条
1 刘晓明;王丽君;赵洋;王昊;;串联故障电弧检测方法的研究[J];电气开关;2014年01期
2 刘晓明;王丽君;侯春光;赵洋;刘湘宁;;基于小波包能量熵的低压串联故障电弧诊断[J];沈阳工业大学学报;2013年06期
3 王子骏;张峰;张士文;顾昊英;曹潘亮;;基于支持向量机的低压串联故障电弧识别方法研究[J];电测与仪表;2013年04期
4 田芳;谌海云;刘丽;卢阿娟;;CENTUM-CS3000系统组态调试及维护[J];仪器仪表用户;2012年06期
5 杨晟健;钟清华;;基于FFT和电磁辐射的低压电弧故障检测[J];现代电子技术;2012年18期
6 刘金琰;栗惠;章建兵;黄兢业;;电弧故障断路器UL标准研究[J];低压电器;2011年18期
7 汪金刚;林伟;王志;李健;何为;王平;;基于紫外检测的开关柜电弧在线检测装置[J];电力系统保护与控制;2011年05期
8 马征;张国钢;柯春俊;;一种基于高频电流频谱分析的故障电弧检测方法[J];低压电器;2010年09期
9 邹云峰;吴为麟;李智勇;;基于自组织映射神经网络的低压故障电弧聚类分析[J];仪器仪表学报;2010年03期
10 杨艺;董爱华;付永丽;;低压故障电弧检测概述[J];低压电器;2009年05期
中国硕士学位论文全文数据库 前7条
1 白静波;基于小波包变换和模糊神经网络的输电线路故障诊断研究[D];太原理工大学;2013年
2 郭家稳;故障电弧模式识别方法的研究[D];沈阳工业大学;2013年
3 桂小智;低压配电系统串联电弧故障实验研究与电弧性短路故障仿真分析[D];重庆大学;2011年
4 郑志成;故障电弧在线诊断技术研究[D];沈阳工业大学;2011年
5 吴建建;航空故障电弧检测技术的研究[D];大连理工大学;2009年
6 杨立树;航空电气系统绝缘故障的研究[D];大连理工大学;2008年
7 吴卓奇;具有电弧检测功能的直流固态功率控制器的研究[D];南京航空航天大学;2008年
,本文编号:2260073
本文链接:https://www.wllwen.com/kejilunwen/hangkongsky/2260073.html