Complete protection scheme for fault detection, classification and location estimation in HVDC transmission lines using support vector machines

被引:81
作者
Johnson, Jenifer Mariam [1 ]
Yadav, Anamika [1 ]
机构
[1] NIT, Elect Engn, GE Rd, Raipur 492010, Chhattisgarh, India
关键词
fault diagnosis; transmission lines; support vector machines; fault location; complete protection scheme; fault detection; fault classification; fault location estimation; HVDC transmission lines; MATLAB; rectifier side AC root mean square voltage; DC voltage; SVM binary classifier; regression algorithm; feature vector;
D O I
10.1049/iet-smt.2016.0244
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
This paper presents a complete protection scheme for detecting, classifying and locating the fault in HVDC transmission lines using support vector machines (SVM). SVM has been used for protective relaying application in HVAC transmission line, however very limited works have been reported for HVDC transmission line. In this work, a +/- 500 kV HVDC transmission system is developed in PSCAD/EMTDC and the measurement signals obtained are analyzed in MATLAB. The rectifier side AC RMS voltage, DC voltage and current on both the poles are continuously monitored, and given as input to the SVM binary classifier in order to detect the presence of fault in the line. Once a fault is detected, the SVM multi-class classification module predicts the type of fault and the SVM regression algorithm predicts the location of fault. The feature vector used in the classification and location modules is the standard deviation of the signals over half cycle before and after the occurrence of fault. The method proposed is simple as it requires single-end data and a direct standard deviation of one cycle data gives very accurate results. The detection and classification modules are 100% accurate whereas the fault location module has a mean error of 0.03%.
引用
收藏
页码:279 / 287
页数:9
相关论文
共 23 条
[1]  
[Anonymous], 2004, Neural Networks, DOI DOI 10.5555/541500
[2]  
Chang C.-C., LIBSVM: a Library for Support Vector Machines
[3]  
Chen P., P 2006 INT C POW SYS, P1
[4]   Practical selection of SVM parameters and noise estimation for SVM regression [J].
Cherkassky, V ;
Ma, YQ .
NEURAL NETWORKS, 2004, 17 (01) :113-126
[5]  
CORTES C, 1995, MACH LEARN, V20, P273, DOI 10.1023/A:1022627411411
[6]   THE APPLICATION OF SATELLITE TIME REFERENCES TO HVDC FAULT LOCATION [J].
DEWE, MB ;
SANKAR, S ;
ARRILLAGA, J .
IEEE TRANSACTIONS ON POWER DELIVERY, 1993, 8 (03) :1295-1302
[7]  
Drucker H, 1997, ADV NEUR IN, V9, P155
[8]   A Novel Fault-Location Method for HVDC Transmission Lines Based on Similarity Measure of Voltage Signals [J].
Farshad, Mohammad ;
Sadeh, Javad .
IEEE TRANSACTIONS ON POWER DELIVERY, 2013, 28 (04) :2483-2490
[9]   A novel whole-line quick-action protection principle for HVDC transmission lines using one-end voltage [J].
Gao, Shu-ping ;
Chu, Xu ;
Shen, Quan-yu ;
Jin, Xing-fu ;
Luo, Juan ;
Yun, Ye-yi ;
Song, Guo-bing .
INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS, 2015, 65 :262-270
[10]   A fault location method based on genetic algorithm for high-voltage direct current transmission line [J].
Li, Yongli ;
Zhang, Shuo ;
Li, Hongbo ;
Zhai, Yongchang ;
Zhang, Weiya ;
Nie, Yao .
EUROPEAN TRANSACTIONS ON ELECTRICAL POWER, 2012, 22 (06) :866-878