Fuzzy Logic Based Fault Classification Scheme for Digital Distance Protection

Khushboo Sahu
2017 International Journal for Research in Applied Science and Engineering Technology  
In this paper, a fuzzy logic-based fault classification scheme for transmission lines is proposed. The classification procedure is carried out by only post fault current phasor of three phases of the transmission line. The proposed technique is able to classify all the possible types of faults including single-phase to ground, two-phases, two-phases to ground and threephase faults with high accuracy. In addition, this method can identify the faulted phase(s) from non-faulted phase(s). The
more » ... phase(s). The proposed method has a good performance in high fault resistances, and high fault distances from relaying point. Large numbers of test cases are generated to verify the performance of proposed technique. The simulation studies have been carried out by using MATLAB/SIMULINK. Sim Power Systems and Fuzzy Logic Toolbox have been used from MATLAB. Keywords-Transmission line, Discrete Fourier transform (DFT), Fuzzy logic system (FLS), Fault detection (FD), Fault classification (FC), Fault inception angle (FIA). I. INTRODUCTION Along with other electrical components, the transmission line suffers from the unexpected failures due to various faults. Protecting of transmission lines is most important task to safeguard electric power systems. For safe operation of transmission line systems, the protection systems should be able to detect, classify, locate accurately and clear the fault as fast as possible to maintain stability in the network. The protective systems are required to prevent the propagation of these faults in the system. The occurrence of any transmission line faults gives rise to the transient condition which may lead to the instability of the system. The purpose of a protective relaying system is to detect all theabnormal signals indicating faults on a transmission system. After detection of the fault, the faulted part should be isolated from the rest of the system to prevent the fault propagation into healthy parts.Transmission line relaying involves three major tasks: fault detection, fault classification and fault location. Fast detection of a transmission line fault enables quick isolation of the faulty line from service and protects it from the transient effects of the fault. Recent protection schemes are based on artificial intelligence (AI) based systems such as artificial neural network (ANN), neurofuzzy and fuzzy logic approaches. Fast and high accurate classification of occurred faults with high reliability is necessary for these techniques because recent fault distance protection schemes utilize the results obtained from fault classification. For example, in ANN-based fault location [1]-[4] and distance protection [5]-[7], the fault classifier performs an important role for enabling the corresponding ANN. Also, the accuracy of fuzzy and fuzzy neural-network-based fault location approaches is highly dependent on the fault classifier operation [8]-[11]. In addition to fault distance location, the ANN and Fuzzy Logic based schemes are also usable in fault classification successfully. The ANN-based approaches are quite accurate in estimating the correct fault type, however, the entire fault and operating conditions such as fault resistance (Rf), fault inception angle (FIA), fault location, system pre-fault load, etc. must be trained for a good performance. Also, ANN has the shortcoming of implicit knowledge representation. On the other hand, the key benefit of fuzzy logic is that the representation of its knowledge is explicit, using simple "If-Then" relations. Also, the fuzzy logic systems are subjective and heuristic and in comparison with ANN, fuzzy-logic techniques can be the simpler and faster methods for planning the fault classifier. In [8] , only the LG (phase to ground) and LLG (two phasesto ground) faults are classified, but the type of fault has notbeen determined. Also, in [1]-[7], [9] and [10] only the natureof the fault (i.e., whether LG, LL, LLG, or symmetrical) hasbeen determined. In [11], all ten types of possible shortcircuitfaults (i.e., a-g, b-g, c-g, a-b, b-c, c-a, a-b-g, b-c-g, c-ag,a-b-c/a-b-c-g) have been determined by using only themagnitude and phase angle of three phase currents.Unfortunately, the proposed fuzzy based strategy in [11] hassome restrictions such as classification errors in high systemloading level, high distances from relaying point and highfault resistance. For circumventing these restrictions, this paper proposes an improved fuzzy logic-based method capable of high accuratefault classification of transmission lines. By using the proposed method, the accuracy of fault classification will considerably increase,
doi:10.22214/ijraset.2017.10099 fatcat:t7fgbe6hafczdlrbqbke7k5zem