• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Society for Machinery Failure Prevention Technology Meeting; 20050418-21; Virginia Beach,VA(US) >FEATURE DIMENSIONALITY REDUCTION FOR PARTIAL DISCHARGE DIAGNOSIS OF AIRCRAFT WIRING
【24h】

FEATURE DIMENSIONALITY REDUCTION FOR PARTIAL DISCHARGE DIAGNOSIS OF AIRCRAFT WIRING

机译：飞机接线局部放电诊断的特征尺寸减小

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Partial discharge (PD) analysis has been successfully used as a means to evaluate the integrity of insulation systems, especially high voltage electrical systems such as generators, transformers, and capacitors. In recent years, PD has also gained increasing attention for its application in low voltage insulation systems such as aircraft wiring systems. Besides detecting and acquiring discharge signals/pulses with sufficiently high measuring sensitivities, one of the challenges of using PD for insulation fault diagnosis is to accurately interpret the acquired signals. That is, to associate the PD signals with real physical phenomena and specifically the states of the insulation that is being monitored. Like in any diagnostic/classification system, the key to an accurate and reliable PD diagnosis is a set of high quality features/attributes that represent/capture the characteristics of PD signals. More importantly, these features must posses strong discriminant power so that the classifier designed based on those features gives desired performance. This paper is concerned with the application of linear feature transformation techniques for reducing the number of features that are extracted by collectively using different feature extraction methods. More specifically, this paper investigates the effectiveness of the two well-studied linear feature transformation methods, namely, principal component analysis (PCA) and linear discriminant analysis (LDA), in improving the classification performance of PD diagnostic systems. We apply these techniques to experimental partial discharge data from an ongoing study of aircraft wiring diagnostics. We also compare the results of this study with those from using other feature selection techniques.
机译：局部放电（PD）分析已成功用作评估绝缘系统（特别是高压电气系统，例如发电机，变压器和电容器）完整性的方法。近年来，PD在低压绝缘系统（例如飞机接线系统）中的应用也引起了越来越多的关注。除了以足够高的测量灵敏度来检测和获取放电信号/脉冲之外，使用PD进行绝缘故障诊断的挑战之一是准确地解释所获取的信号。也就是说，将PD信号与实际的物理现象相关联，尤其是将其与正在监视的绝缘状态相关联。像在任何诊断/分类系统中一样，进行准确可靠的PD诊断的关键是代表/捕获PD信号特征的一组高质量特征/属性。更重要的是，这些功能必须具有强大的判别能力，以便基于这些功能设计的分类器具有理想的性能。本文涉及线性特征变换技术在减少通过共同使用不同特征提取方法共同提取的特征数量方面的应用。更具体地说，本文研究了两种经过深入研究的线性特征转换方法（主成分分析（PCA）和线性判别分析（LDA））在提高PD诊断系统的分类性能方面的有效性。我们将这些技术应用于正在进行的飞机布线诊断研究中的局部放电实验数据。我们还将本研究的结果与使用其他特征选择技术的结果进行比较。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号