A power system restoration method using voltage source converter–high-voltage direct current technology, aided by time-series neural network with firefly algorithm

Abstract

Voltage source converter (VSC)–high-voltage direct current (HVDC) has become a new trend for achieving an efficient and reliable bulk power transmission. The main challenges in VSC-HVDC link are the “soft-start-up” or the restoration of power after the system blackout, with over voltage and less response delay. In this paper, VSC-HVDC link with firefly Levenberg–Marquardt-trained artificial neural network (ANN), termed as ANN-FLM, is proposed to restore the power quickly. The simulation is performed by creating blackout in the system through the introduction of a three-phase fault. The performance analysis is made on VSC-HVDC with ANN-FLM in existing proportional integral controller. The simulated results provide faster restoration and independent control of real and reactive powers. The proposed method minimizes the over voltage to 98% in maximum point, 84% in phase A, 61% in phase B and 82% in phase C, when simulated.