基于BP网络的GIS局部放电声电联合检测故障定位方法

摘要

如何快速定位GIS放电位置，是局部放电在线检测的关键。基于超声波和超高频多传感器信息融合局部放电联合定位方法，是将来自放电位置这一目标的多源信息加以智能合成。声电联合定位检测系统采用到达时间差TDOA法，利用BP神经网络对超声波和超高频法的传感器采集数据进行融合。最后通过仿真验证，得出其定位准确率较单一方法大大提高。%The key of on-line partial discharge detection is how to quickly locate the fault location. Method of partial discharge acoustic electric combined based on ultrasonic and ultra high frequency multi sensor information fusion, to make intelligent synthesis of multi source information from the discharge po-sition. In this paper, the structure of acoustic electric joint position detection system is designed, and TDOA method is used to get the time difference, BP neural network is used to fuse the data of ultrasonic wave and ultra high frequency sensor. Case simulation shows that, the positioning accuracy greatly im-proved.