Detection and Classification of Transmission Line Faults Using Empirical Mode Decomposition and Rule Based Decision Tree Based Algorithm

The work presented in this paper is mainly focused to develop a technique based on the Hilbert Huang Transform (HHT) and rule based decision tree (RBDT) for detection and classification of the power system faults. The Hilbert Huang Transform is used in two parts known as Hilbert Transform (HT) and Empirical Mode Decomposition (EMD) for detection of faults in power system network. A test system comprising of a transmission line connected at both ends to AC sources is used for the proposed study. Current is recorded at both ends of the transmission line. Current is decomposed using the Hilbert Transform and absolute value of output is obtained. This output is decomposed using the empirical mode decomposition and intrinsic mode functions (IMF) are obtained at first and second level of decomposition. A fault index has been proposed to detect the various types of the faults and it is obtained by multiplying the IMFs at level 1and 2. The threshold values of the proposed fault index are given as input to the rule based decision tree for classification of the faults. The results of the proposed method are also compared with a method already reported in the literature. The investigated faults include the line to ground (LG), double line (LL), double line to ground (LLG) and three phase fault involving ground (LLLG). It is established that the proposed approach effectively detects and classify the power system faults.