基于特高频无线传感器和模式识别算法的局部放电定位法  

Partial Discharge Localization Methodology Based on Ultra High Frequency Wireless Sensor and Pattern Recognition Algorithm

作　　者：黄辉 黄凤 李春龙 李臻[2] 

机构地区：[1]全球能源互联网研究院,南京210000 [2]上海交通大学智能输配电研究所,上海200240

出　　处：《科学技术与工程》2018年第4期65-70,共6页Science Technology and Engineering

基　　金：2016年国家电网公司科技项目资助

摘　　要：对电力设备局部放电的检测与定位是保障电力系统安全稳定运行的重要手段之一。现有局部放电定位法主要是基于特高频传感器技术和时差法进行的,高的采样率和同步精度,使得其硬件成本巨大、实现困难,且容易受现场环境影响。提出了基于特高频无线传感器和模式识别算法的局部放电定位法,该方法硬件要求低,易于实现,且具有良好的环境适应性。首先通过现场测量,建立待检测区域局部放电信号强度与放电坐标的特征信息库。当有局部放电发生时,将此时传感器测量到的特征信息输入已建好的信息库中进行模式识别,从而得到定位结果。现场试验结果表明,提出的新型局部放电定位算法的平均定位误差为0.58 m,80.8%的定位误差小于1 m,从而验证了算法的有效性,具有较好的推广应用价值。The detection and localization of partial discharge is one of the important method for the security and stability of power systems. Exist partial discharge localization methods are generally based on UHF technology and time difference method which inquiring a high speed sampling rate and synchronization accuracy. The high hardware cost and vulnerability to the environmental influence make it difficult to widely application. A discharge localization method based on UHF wireless sensor and pattern recognition is proposed,which is easy to implement and has good environmental adaptability. Firstly,the partial discharge signal strength and discharge coordinates information are established by site survey. When a partial discharge happens,the UHF strength are measured by the sensors and input into the prebuilt library for pattern recognition to obtain the PD location. The field test showed that the mean errors of PD localization by proposed method is 0. 58 m,and 80. 8% of errors less than 1 m which proved the effectiveness of the algorithm proposed.

关 键 词：局部放电 定位算法 模式识别 神经网络 

分 类 号：TM835.1[电气工程—高电压与绝缘技术]