Leakage Line Detection of Coal Mine Low Voltage Distribution Network with Evidence Theory Fusion Correlations

• Published in: 2010 International Conference on Measuring Technology and Mechatronics Automation Vol. 2; pp. 583 - 586
• Main Authors: Jianwen Zhao, Yuanbin Hou, Shangbin Sun, Xuejie Cheng
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.03.2010
• Subjects: Automation; coal mine leakage; evidence theory; fusion; Leak detection; Low voltage; Mechatronics; multi correlation; signal distance
• ISBN: 1424450012; 9781424450015
• ISSN: 2157-1473
• DOI: 10.1109/ICMTMA.2010.191

In order to meet rapidity and reliability of leakage protection, transient zero-sequence current wave form difference characteristics between leakage line and normal line is give through theory analysis; A novel leakage line detection method using multi-correlation criterion fusion is presented. According zero sequence wave form characteristics, by D-S evidence theory, from making of leakage measures, discrimination frame, probability assignment function, leakage judgment formula, the multi criterion leakage line detection is changed into evidence reasoning. Through making fusion of signal distance, correlation coefficient and gray absolute relation index, and then this method is established. Simulation model of 660 V coal mine network is set up using MATLAB, to verify the validity of this leakage line detection method. Theory analysis and simulation indicate that this method makes the difference between leakage line and normal line more distinct than single leakage line detection method; algorithm precision is not influenced by different lines, leakage resistances, leakage angles, and leakage locations; datum time range is about 10 ms, leakage line detection is rapidly.