基于大数据的配网告警信号中关键信号的自动识别研究

详细信息    查看全文 | 推荐本文 |

英文篇名：Research on automatic identification of key signals in distribution network alarm signals based on big data 作者：刘远龙 ; 潘筠 ; 潘颖 ; 刘帅 英文作者：LIU Yuanlong;PAN Yun;PAN Ying;LIU Shuai;Shandong Electric Power Co.,Ltd.;Shandong Power Company Weihai Power supply Company; 关键词：大数据 ; 告警信号 ; 信号识别 英文关键词：big Data;;alarm signal;;signal recognition 中文刊名：ZDYY 英文刊名：Automation & Instrumentation 机构：国网山东省电力公司;国网山东省电力公司威海供电公司; 出版日期：2019-04-25 出版单位：自动化与仪器仪表 年：2019 期：No.234 语种：中文; 页：ZDYY201904055 页数：5 CN：04 ISSN：50-1066/TP 分类号：226-230

摘要

为了准确地识别出配网告警信号中存在的关键信号,需要对配网告警信号中关键信号的识别方法进行研究。采用当前识别方法对配网告警信号中的关键信号进行识别时,存在识别结果精准度低和识别效率低的问题。提出一种基于大数据的配网告警信号中关键信号的自动识别方法,采用EMD分解方法分解配网系统中存在的告警信号,并对告警信号进行Hilbert变化,得到告警信号的边际谱和Hilbert谱,通过分析边际谱和Hilbert谱分别得到告警信号的真实频率、信号特征和故障时刻。采用时域信息熵测度方法计算采集数据窗中存在的能量信息熵测度值,根据计算结果得到信息熵测度值的标准差,以标准差为依据完成配网告警信号中关键信号的识别。实验结果表明,所提方法识别结果的精准度高、识别效率高。
        In order to accurately identify the key signals existing in the distribution network alarm signal,it is necessary to study the identification method of the key signals in the distribution network alarm signal.When the current identification method is used to identify the key signals in the distribution network alarm signal,there is a problem that the recognition result is low in accuracy and the recognition efficiency is low.An automatic identification method for key signals in distribution network alarm signals based on big data is proposed.The EMD decomposition method is used to decompose the alarm signals existing in the distribution network system,and the Hilbert change is performed on the alarm signals to obtain the marginal spectrum and Hilbert spectrum of the alarm signals.By analyzing the marginal spectrum and the Hilbert spectrum,the real frequency and signal characteristics of the alarm signal and the time of failure are obtained.The time domain information entropy measure method is used to calculate the energy information entropy measure value existing in the data window.According to the calculation result,the standard deviation of the information entropy measure value is obtained,and the key signal in the distribution network alarm signal is identified based on the standard deviation.The experimental results show that the proposed method has high accuracy and high recognition efficiency.

引文

[1] 怀全.基于PLC的大型电气工程设备故障自动诊断系统设计[J].自动化与仪器仪表,2017,(12):127-129.
    [2] 李莹莹,焦超群,王飞.基于Labview的超声局部放电检测仪声音模块设计[J].电子设计工程,2017,25(5):165-169.
    [3] 苑亚南,朱希安,张涛.电机噪声故障信号优化检测仿真研究[J].计算机仿真,2017,34(10):401-405.
    [4] 贺利芳,崔莹莹,张天骐,等.基于幂函数型双稳随机共振的故障信号检测方法[J].仪器仪表学报,2016,37(7):1457-1467.
    [5] 朱明斯,朱希安.基于新阈值函数的电动机故障信号小波去噪[J].科技通报,2017,33(8):100-104.
    [6] 岳应娟,孙钢,蔡艳平,等.基于RP与NMF的内燃机气阀故障诊断方法[J].科学技术与工程,2016,16(28):85-89.
    [7] 张艳霞,尹佳鑫,蒙高鹏,等.基于错误逻辑区域检测的配电网容错故障定位[J].电机与控制学报,2017,21(11):1-8.
    [8] 雷勇,何晓,周聪聪.一种单端注入-多端检测的配电网故障分支判定方法[J].电网技术,2016,40(10):3243-3249.
    [9] 崔坤利,郎朗,陈孟元,等.配电网中基于分簇定位的WSN节点故障定位研究[J].传感技术学报,2017,30(1):146-151.
    [10] 陈长浩,王安.一种便携式信号电缆接地故障检测仪[J].计算机测量与控制,2017,25(7):317-319.
    [11] 崔伟成,李伟,孟凡磊,等.基于果蝇优化算法的自适应随机共振轴承故障信号检测方法[J].振动与冲击,2016,35(10):96-100.
    [12] 亓晓燕,刘恒杰,王函旭,等.基于小波理论的电力系统故障信号识别研究[J].自动化与仪器仪表,2017,(8):4-6.
    [13] 周伟,王甜甜.故障信号检测虚拟示波器数据采集系统设计[J].现代电子技术,2016,39(20):97-100.
    [14] 党建,何洋洋,贾嵘,等.水轮发电机组非平稳振动信号的检测与故障诊断[J].水利学报,2016,47(2):173-179.
    [15] 梁辰,邰能灵,胡炎,等.基于保护信号的隐性故障检测[J].电力自动化设备,2016,36(5):154-159.