Presenting a hybrid method for fault tolerance based on neural fuzzy logic in distribution networks using phasor measurement units
Pengwei Zhang,Yiying Tu,Yonggang Zeng,Qun Yi
DOI: https://doi.org/10.1007/s12652-024-04876-x
IF: 3.662
2024-10-10
Journal of Ambient Intelligence and Humanized Computing
Abstract:In this paper, a hybrid approach to increase the performance reliability of power distribution networks using phasor measurement units (PMU) is presented. Electricity distribution networks are very important as a vital part of electricity transmission systems in providing energy to users. Considering the ever-increasing complexity and the huge amount of existing demand, maintaining the optimal performance and reliability of these networks is vital. One of the main challenges in this industry is identifying and finding problems. In this regard, improvements in phasor measurement technology using phasor measurement units (PMU) allow engineers in this industry to more accurately evaluate data, diagnose and locate network faults. In this research, one of the non-linear methods for finding ground faults in power distribution networks using voltage phasor measurement in several network stations using phasor measurement units (D-PMU) has been demonstrated. In the first approach, genetic optimization algorithms and bullet optimization algorithm (PSO) have been used for nonlinear modeling of fault position estimation based on different types of 1-phase, 2-phase and 3-phase faults. The second method uses fuzzy network training to provide details about phasor voltages and fault types. By simulating a 9-station system using MATLAB software, the usefulness of the proposed methods has been shown. In modeling, 1-phase, 2-phase and 3-phase faults along with different line lengths and line characteristics at different stations have been investigated. Also, the findings are presented and the location of the defect is identified immediately.
computer science, information systems,telecommunications, artificial intelligence