A faulty feeder selection method based on the slope of zero sequence current fitting curve for the non-effectively grounded system

When a single-phase grounded (SPG) fault occurs in the distribution network, it is essential to ensure the dis-tribution network's safe and reliable operation to select faulty feeders. This paper analyzes the slope of the zero sequence current fitting curve of each feeder when an SPG fault occurs in the distribution network. According to the different slope values between the faulty feeder and the healthy feeder at the zero sequence current mutation time, this paper proposed an SPG faulty feeder selection method based on the slope of the zero sequence current fitting curve. A locally weighted linear regression (LWLR) algorithm is first used to fit the zero sequence current sampled data. Then, the mathematical expression of the zero sequence current fitting curve is derived, and the corresponding sloping curve is obtained. Finally, the slope at the zero sequence current mutation time is used to determine the faulty feeder. If the slope value of all fitting curves at the zero sequence current mutation time is negative, the bus fault is determined; otherwise, the feeder with a positive slope value of the fitting curve is determined as the faulty feeder. In this method, the power frequency component of zero sequence current does not need to be filtered out, making it easy to achieve good detection sensitivity in high-resistance grounded fault selection. MATLAB/Simulink simulation results indicate that the proposed method can detect the SPG fault accurately under various fault conditions.