A new differential protection algorithm for power reactors based on the second central moment. (June 2020)

• Record Type: Journal Article
• Authors: Vázquez, Ernesto; Andrade, Manuel A.; Esponda, Héctor; Ávila, Jesús
• Is Part Of: International journal of electrical power & energy systems. Volume 118(2020)
• Publication Date: 2020-06
• Subjects: Differential protection -- Shunt reactor -- Inrush current -- Second central moment -- PSCAD; Electrical engineering -- Periodicals; Electric power systems -- Periodicals; Électrotechnique -- Périodiques; Réseaux électriques (Énergie) -- Périodiques; Electric power systems; Electrical engineering; Periodicals; 621.3
• DOI: 10.1016/j.ijepes.2019.105795
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: The method is not affected by transient events of the power reactor such as inrush current, CT saturation and external faults. The method is setting-free, and it does not depend on the reactor parameters or the network configuration. The method is capable to detect turn-to-turn fault with at least a 2% of short-circuited turns. The fault detection time is less than one cycle of 60 Hz. Abstract: Power shunt reactors are critical elements in electric power systems for controlling reactive power and maintaining voltage within a permissible range in transmission systems. Protection schemes for shunt reactors are similar to those used in power transformers, e.g., the differential, overcurrent, and distance protection schemes. However, these protection schemes may misoperate for several reasons such as a decaying DC component when the reactor is energized, or the occurrence of current transformer saturation when an external fault occurs. In this context, a novel differential protection algorithm for power shunt reactors is proposed. This algorithm uses the magnitude of the second central moment as a basis for identifying whether the event corresponds to an internal fault. The proposed algorithm was tested and implemented using PSCAD and the MATLAB software. Current transformer saturation and noise were considered in all the events evaluated. The results show that the proposed algorithm accurately discriminates between internal and external faults as well as transient events such as inrush currents. Furthermore, the algorithm exhibits a limited capability for detecting turn-to-turn faults until 2% of short-circuited turns.