| 基于小波特征的油浸式变压器电弧光故障诊断 |
| |
| 引用本文： | 李昊宇,李建兴,马莹,罗堪.基于小波特征的油浸式变压器电弧光故障诊断[J].福建工程学院学报,2019,0(1):72-76. |
| |
| 作者姓名： | 李昊宇 李建兴 马莹 罗堪 |
| |
| 作者单位： | 福建工程学院 信息科学与工程学院 |
| |
| 摘 要： | 基于小波分解频带能量特征和BP神经网络的方法识别油浸式变压器短路故障。利用电弧光信号进行油浸式变压器短路故障诊断，对不同工况下的光信号进行多分辨率分析的四层小波分解，选择合适的重构小波系数，提取特征频带。对提取出的特征频带的小波系数作平方和归一化处理，求出每个特征频带的能量，作为特征参数输入到BP神经网络模型中进行训练和故障识别。
|
| 关 键 词： | 电弧光 小波频带能量 BP神经网络 故障识别 |
Fault diagnosis of arc light of oil-immersed transformer based on wavelet characteristics |
| |
| LI Haoyu,LI Jianxing,MA Ying,LUO Kan.Fault diagnosis of arc light of oil-immersed transformer based on wavelet characteristics[J].Journal of Fujian University of Technology,2019,0(1):72-76. |
| |
| Authors: | LI Haoyu LI Jianxing MA Ying LUO Kan |
| |
| Institution: | School of Information Science and Engineering, Fujian University of Technology |
| |
| Abstract: | Short circuit fault detection of oil-immersed transformers was studied based on wavelet decomposition band energy characteristics and BP neural network. The short-circuit fault diagnosis of oil-immersed transformer was carried out by using arc light signals, four-layer wavelet decomposition of multi-resolution analysis of optical signals under different working conditions was performed. Appropriate reconstructed wavelet coefficients were selected to extract characteristic bands. The extracted wavelet coefficients of the characteristic frequency band were normalized by the sum of squares, obtaining the energy of each characteristic frequency band, which was input as a characteristic parameter into the BP neural network model for training and fault detection. |
| |
| Keywords: | arc light wavelet band energy BP neural network fault identification |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《福建工程学院学报》浏览原始摘要信息 |
| 点击此处可从《福建工程学院学报》下载免费的PDF全文 |
