SVM based zonal setting of Mho relay for shunt compensated transmission line

• Impact of Shunt FACTS devices such as STATCOM and SVC on protection of transmission line is discussed.
• Equivalent impedance inserted by the device in transmission line is calculated in terms of current injected by the device.
• Actual fault impedance is calculated after removing the error introduced by shunt FACTS device.
• A SVM zonal classifier is generated to classify fault in between zone 1 and zone 2.

Presence of shunt FACTS (Flexible Alternating Current Transmission System) device significantly affect the performance of protection system and may create security and reliability issues. This paper introduces a novel approach for zone identification in transmission line employing shunt FACTS devices such as Static Var Compensator (SVC) and Static Synchronous Compensator (STATCOM). The technique depends on the modified apparent impedance seen by the impedance relay for all possible operating conditions. In this technique the Support Vector Machine (SVM) is used for discriminating faulty zone (zone 1 or zone 2). In addition an optimization technique viz. Genetic Algorithm (GA) is used to optimize SVM parameters. A typical 230 kV system was simulated in PSCAD/EMTP software and the results show that the proposed scheme is secure, accurate and reliable under the wide variation in power system parameters like load angle, fault resistance, fault location, inception angle and compensation level.