Title: Fault optimisation on IEEE 14-bus system with machine learning-based SPLIT TCSC

Authors: Niharika Agrawal; Faheem Ahmed Khan; Mamatha Gowda

Addresses: Department of Electrical and Electronics Engineering, Ghousia College of Engineering, Ramanagaram District – 562 159, Karnataka, India ' Department of Electrical and Electronics Engineering, Ghousia College of Engineering, Ramanagaram District – 562 159, Karnataka, India ' Department of Artificial Intelligence and Data Science, BGS College of Engineering and Technology, Mahalakshmi Puram, Bengaluru, 560 086, Karnataka, India

Abstract: All the operations of the power system should be smooth, continuous, reliable, and efficient. An electric fault interrupts the normal flow of power, damages the electrical equipment, creates abnormal voltages and currents, and deteriorates the system's power quality. In this manuscript, a novel device, the SPLIT TCSC, is implemented with the traditional proportional-integral (PI) controller and with controllers based on machine learning algorithms such as artificial neural network, and random forest for fault optimisation on IEEE 14-bus system. These controllers showed better results than the PI controller with respect to voltage during fault time and THD reduction. The voltage drop is around 30 kV, and the THD is around 1.95% with the system based on random forest. The results showed that power is utilised more effectively with the application of these algorithms in the system, along with improvements in the power quality, reliability, and efficiency of the system.

Keywords: continuous; control; current; efficiency; fault; optimisation; oscillations; power; reliable; supply; voltage.

DOI: 10.1504/IJPEC.2022.130964

International Journal of Power and Energy Conversion, 2022 Vol.13 No.3/4, pp.319 - 347

Received: 03 Aug 2022
Accepted: 13 Mar 2023
Published online: 14 May 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article