摘要
配电网中的单相接地故障是经常发生的一类故障,文章通过设计研究并安装一款基于神经网络数据融合方法的小电流选线装置来排查此类故障。装置利用BP神经网络良好的非线性拟合能力,结合故障后各线路的小波能量、五次谐波和有功功率作为特征值进行融合选线。根据故障选线的准确性和快速性对系统的要求,提出了一种基于ARM(Advanced RISC Machines)技术的软硬件平台,装置采用32位嵌入式ARM微处理器S3C2440作为诊断机的核心,以C8051F060作为前端采集器的控制核心,充分利用ARM出色的实时中断响应和快速的运算速度,很好的实现了综合智能选线算法。现场运行结果表明,该装置运行稳定,选线快速可靠,适应性强,是小电流接地系统中一种实用的选线装置。
Single-phase grounding fault frequently occurs in distribution network .In this paper , the design and instal-lation of a new small current grounding line selection device is used to trouble shoot such fault .The device uses the nonlinear fitting ability of BP neural network , combining with wavelet energy , fifth harmonic and active power to de-termine the fault line .A design of hardware and software platform on ARM （ Advanced RISC Machines ） meeting to the requirement of reliability and rapidity in fault line detection is presented .ARM processor S3C2440 makes the con-trolling core of diagnostic machine and C 8051 F060 makes the controlling core of the data acquisition unit .The advan-tages of ARM , excellent real time interrupt response and high calculating speed , make the synthetical intelligent algo-rithm to be fast processed in time .The operation test shows that the device has good running stability and the fault de-tection is fast and reliable .It is a practical line selection device .
出处
《电测与仪表》
北大核心
2015年第18期74-79,共6页
Electrical Measurement & Instrumentation
关键词
配电网
小电流接地故障
数据融合
神经网络
ARM
故障选线
distribution network, small current grouriding fauh, data fusion, neural network, ARM,fault line detection
