Estimation of arc fault location in power transmission lines with travelling wave methods

With this thesis, arc fault in Power Transmission Lines (PTL) are simulated by transferring arc model in simulation environment and arc fault location is estimated by Travelling Wave (TW) methods. Transient Frequency Spectrum Analysis (TFSA) that was presented before for normal fault location is presented here with fault detection, data size restriction and a new way of reducing source impedance effect improvements. Approximate Derivative (AD) that was presented with double end synchronized data is presented here for single end data for the first time. With the aid of TW theory, reflected TWs are distinguished based on sign change with a comprehensive algorithm. AD is also suggested for fault detection. With restricted data needed through fault detection, fault location is estimated with high accuracy. Restricted data with the aid of fault detection ensures both methods to give results faster. In this way, reducing destructive effects of arc is aimed. It is proven that both methods presented are not affected by the nonlinear characteristics of the arc, and their reactions to different cases in power systems are examined. Besides, the effect of noise contemned in the signals on methods are investigated. As a result, the effectiveness and originality of the methods are proven by comparison with previous studies. Simulations are made in ATP/EMTP environment, fault detection and fault location calculations are made in MATLAB.