Dynamic Voltage Restorer With Quasi-Newton Filter-Based Control Algorithm and Optimized Values of PI Regulator Gains

In this paper, a three-phase three-wire dynamic voltage restorer (DVR) system has been chosen with quasi-Newton (QN) filter-based control algorithm. It uses Hessian matrix iteratively which is also named as QN based on modified Broyden- Fletcher-Goldfarb-Shanno (MBFGS) in each phase. The main features of this algorithm are the fundamental frequency estimation, rejection of harmonics its components, and tracking of the peak value of voltage or current. As DVR in this application has been implemented for compensation of voltage disturbances like imbalance and distortions along with voltage sag and swell, the fundamental component (FC) of the supply voltage is essential components to design control algorithm. The QN technique using MBFGS has been used for exaction of the FC from each disturbed supply voltage, which has been utilized in generating the reference load voltage. For proportional integrator (PI) regulator gains estimation, the population-based optimization technique, called as multiverse optimization (MVO), has been introduced in this paper. The MVO algorithm has adaptive coefficients which gives the benefits of smoothly maximizing the possibility of wormholes in the universe and increases the accuracy of the local search. The validation of control algorithm with optimization techniques-based PI tuning for DVR has been done using MATLAB software and real-time implementation.