A Curve Fitting–Based Directional Protection of Transmission Lines Using Genetic Programing

Abstract

In modern power systems, there is a growing need for protection techniques that are fast, precise, and stable to enhance the reliability and stability of power networks. This paper introduces an advanced technique for fault direction estimation that addresses key limitations in existing methods. Our proposed algorithm eliminates the dead‐zone issue, effectively handling both forward and reverse faults regardless of their location, including faults near relay regions. Using genetic programing (GP), we derive a closed‐form equation to drive the fault direction estimation algorithm, ensuring accurate and reliable results. The algorithm demonstrates exceptional speed and reliability in fault detection and direction estimation, even under challenging conditions such as current transformer (CT) saturation, power swing, and source strength variations. It is also insensitive to fault resistance and fault inception angle, increasing its robustness. A notable advantage of this algorithm is its efficiency at low sampling frequencies, making it highly adaptable for industrial applications and compatible with modern software and hardware requirements. We validated the algorithm through simulations on the electromagnetic transients program (EMTP‐RV) platform and with field data from the Manesht 230 kV substation in Ilam province, Iran. The results confirm the algorithm’s impressive speed and high reliability.