基于置信网络的直流输电线路暂态信号识别方法Transient Signal Identification Method for DC Transmission Line Based on Belief Network
姚长元,罗国敏
YAO Changyuan,LUO Guomin
摘要(Abstract)：
暂态保护是高压直流输电系统的主保护,提高暂态信号的识别精度是改善保护性能的有力措施。单极接地故障、雷击干扰和雷击故障是高压直流输电线路常见的3类暂态信号。提出一种基于置信网络的暂态信号识别方法,以原始数据作为输入,利用网络实现特征提取和识别。该方法减少了复杂的信号处理及特征提取环节,能够对多种场景下的暂态信号进行有效识别。仿真结果表明,该方法在低采样频率、短时间窗和不同系统参数的条件下,对三类暂态信号具有良好的识别效果。
Transient protection is the primary protection of HVDC transmission system. Improving the identification accuracy of transient signals is a powerful measure to improve protection performance. Single-pole grounding fault, lightning disturbance and lightning fault are three common types of transient signals in HVDC transmission lines. This paper proposes a transient signal identification method based on belief network, which uses raw data as the input and uses network to for feature extraction and identification. The method reduces complicated signal processing and feature extractions, and can effectively identify transient signals in various scenarios. As demonstrated by simulation results, under the conditions of low sampling frequency, short time window and different system parameters, it has a good identification effect on three types of transient signals.
关键词(KeyWords)：
高压直流输电;信号识别;置信网络
HVDC transmission;signal identification;belief network
基金项目(Foundation):
作者(Author):
姚长元,罗国敏
YAO Changyuan,LUO Guomin
DOI: 10.19585/j.zjdl.201906001
参考文献(References)：
- [1]汤广福,贺之渊,庞辉.柔性直流输电工程技术研究、应用及发展[J].电力系统自动化,2013,37(15):3-14.
- [2]汤广福,罗湘,魏晓光.多端直流输电与直流电网技术[J].中国电机工程学报,2013,33(10):8-17.
- [3]梁旭明,张平,常勇.高压直流输电技术现状及发展前景[J].电网技术,2012,36(4):1-9.
- [4]宋国兵,高淑萍,蔡新雷,等.高压直流输电线路继电保护技术综述[J].电力系统自动化,2012,36(22):123-129.
- [5]高锡明,张鹏,贺智.直流输电线路保护行为分析[J].电力系统自动化,2005(14):96-99.
- [6]陶瑜,龙英,韩伟.高压直流输电控制保护技术的发展与现状[J].高电压技术,2004(11):8-10.
- [7]王钢,李海锋,赵建仓,等.基于小波多尺度分析的输电线路直击雷暂态识别[J].中国电机工程学报,2004(4):143-148.
- [8]何正友,蔡玉梅,钱清泉.小波熵理论及其在电力系统故障检测中的应用研究[J].中国电机工程学报,2005(5):40-45.
- [9]陶瑜,龙英,韩伟.高压直流输电控制保护技术的发展与现状[J].高电压技术,2004(11):8-10.
- [10]王钢,李海锋,赵建仓,等.基于小波多尺度分析的输电线路直击雷暂态识别[J].中国电机工程学报,2004(4):143-148.
- [11]HINTON G E,SALAKHUTDINOV R R.Reducing the dimensionality of data with neural networks[J].Science,2006,313(5786):504-507.
- [12]DENG L,YU D.Deep learning:Methods and applications[J].Foundations and Trends in Signal Processing,2014,7(3/4):197-387.
- [13]徐政,薛英林,张哲任.大容量架空线柔性直流输电关键技术及前景展望[J].中国电机工程学报,2014,34(29):5051-5062.
- [14]刘剑,邰能灵,范春菊,等.多端VSC-HVDC直流线路故障限流及限流特性分析[J].中国电机工程学报,2016,36(19):5122-5133.
- [15]文远芳.高电压技术[M].武汉:华中科技出版社,2000.
- [16]HINTON G E.A Practical Guide to Training Restricted Boltzmann Machines[M]//Neural Networks:Tricks of the Trade.Springer Berlin Heidelberg,2012:599-619.
- [17]HINTON G E,OSINDERO S,TEH Y W.A fast learning algorithm for deep belief nets[J].Neural computation,2006,18(7):1527-1554.