Jeevan J. Inamdar
School of Electrical Engineering, Vellore Institute of Technology, Chennai, Tamil Nadu, India.

K. Iyswarya Annapoorani
Centre for e-Automation Technologies, School of Electrical Engineering, Vellore Institute of Technology, Chennai, Tamil Nadu, India.

DOI https://doi.org/10.33889/IJMEMS.2025.10.6.085

  ;

Abstract

The power quality of the electrical supply system has been impacted by the increase in non-linear loads and renewable energy sources such as the solar and the wind power integrated into the grid. The poor power quality results in equipment damage, system shutdown, and data loss, leading to lower operational efficiency and higher maintenance costs. The voltage sag, voltage swell, voltage flicker and voltage harmonic distortions are the most common and severe issues. The dynamic voltage restorer is the most effective solution for these voltage quality problems, which injects necessary voltage levels at the time of faults in order to maintain the load side voltage within specified boundaries. The appropriate control technique should be adopted for the generation of firing pulses for power electronic switches used to construct this device. The classical sliding mode control has the disadvantage of taking a long time to minimize errors. The main objective of this paper is to improve the accuracy and durability the conventional algorithm by adding nonlinear quantity, resulting in the fast terminal sliding mode controller. Furthermore, the artificial neural network is combined with proposed methodology to improve performance of double feeder power system for nonlinear load conditions. The system is modeled and simulated in a MATLAB/Simulink environment with a proportional integral controller optimized by particle swarm optimization, genetic algorithm and mind blast algorithm followed by non-linear controllers. The faults are simulated to mimic voltage sags A, E and B with different load conditions like linear, dynamic and non-linear. The power quality indices like total harmonic distortion, harmonic compensation ratio, sag score and voltage sag lost energy index are considered for assessment of compensation percentage. It is observed that Sag Score is improved by 80% thus increasing voltage sag lost energy index to more than 95%. Therefore, quantitative data demonstrate the efficacy of the proposed method in mitigating voltage sag while simultaneously reducing grid voltage imbalance and distortion, irrespective of the fault type.

Keywords- Power quality, Fault conditions, Dynamic voltage restorer, Sliding mode control, Artificial neural network.

Citation

J. Inamdar, J., & Annapoorani, K. I (2025). Voltage Quality Analysis using Dynamic Voltage Restorer Based on Non-Linear Control Algorithms. International Journal of Mathematical, Engineering and Management Sciences, 10(6), 1824-1859. https://doi.org/10.33889/IJMEMS.2025.10.6.085.