• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Second ICSC Symposium on Engineering of Intelligent Systems, Jun 27-30, 2000, Scotland, U.K. >DIELECTRIC CONDITION MONITORING OF HIGH VOLTAGE SYSTEMS USING INTELLIGENT PARTIAL DISCHARGE ANALYSIS
【24h】

DIELECTRIC CONDITION MONITORING OF HIGH VOLTAGE SYSTEMS USING INTELLIGENT PARTIAL DISCHARGE ANALYSIS

机译：基于智能局部放电分析的高压系统介电状态监测

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

Dielectric materials forming the insulation of high-voltage plant can take a number of different forms dependant upon the criteria an application duly demands. The condition of a dielectric is of extreme importance to maintain supply stability and safety, thus any degradation must be identified and rectified at an early stage to continue an operational duty. Many factors of both extraneous and inherent natures, affect service performance, thus methods providing early diagnosis of dielectric faults, would greatly assist plant-replacement and maintenance programs. An area which research has linked to premature ageing and reduced service performance, is that of partial discharge phenomena (pd). As inevitable inclusions, small voids within dielectric masses and similarly contaminants within insulating liquids, introduce areas of differing dielectric stress remaining at all times prone to pd. The discharge captures evident of such deterioration, are attributable to the very components of origin and thus illustrative of root. By conducting an analysis of pd behaviour and collating historical behaviour patterns, an accurate assessment of the dielectric condition and a subsequent remaining life prediction may be deduced. Neural networks have previously been exploited in investigations as trainable pattern classifiers, where there has been a need to characterise a voltage, or current signature with a physical phenomena. Presented with representative and characteristic signatures, neural networks provide potential methods by which partial discharges may be collated and categorised against factors of source. The subject of this paper is the combination of partial discharge techniques with Neural Network analysis, to form a powerful method for the diagnosis of faults in dielectrics. This paper describes the latest work in this area which is being carried out in the Engineering Research Centre of the University of Brighton and includes the results of experimental investigations which have been carried out there.
机译：取决于应用程序适当要求的标准，形成高压设备绝缘的介电材料可以采用多种不同的形式。电介质的状况对于保持电源的稳定性和安全性至关重要，因此必须尽早识别并纠正任何劣化，以继续工作。许多无关紧要的因素和固有的因素都会影响服务性能，因此提早诊断介电故障的方法将极大地帮助工厂更换和维护程序。与过早老化和服务性能下降相关的研究领域是局部放电现象（pd）。作为不可避免的夹杂物，电介质块中的小空隙以及绝缘液体中的类似污染物会引入始终存在pd的不同介电应力区域。放电记录表明了这种劣化，这归因于原产地的组成部分，因此可以说明根。通过对钯行为进行分析并核对历史行为模式，可以得出介电条件的准确评估以及随后的剩余寿命预测。神经网络先前已在研究中用作可训练的模式分类器，其中需要用物理现象来表征电压或电流特征。具有代表性和特征性签名的神经网络提供了可能的方法，通过这些方法可以根据来源因素对局部放电进行整理和分类。本文的主题是将局部放电技术与神经网络分析相结合，以形成一种诊断电介质故障的有效方法。本文介绍了在布莱顿大学工程研究中心正在进行的这一领域的最新工作，并包括了在那里进行的实验研究的结果。

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