Implementation Of Conjugate Gradient Backpropagation Neural Network Control Algorithm for Single-Phase Grid Tied SPV-DSTATCOM System

Nor Hanisah Baharudin; Aini Syahida Mohd Shaizad; Tunku Muhammad Nizar Tunku Mansur; Rosnazri Ali; Mohd Syahril Noor Shah; Mohd Zamri Hassan

doi:10.37934/ctds.6.1.113

Authors

Nor Hanisah Baharudin Centre of Excellence for Renewable Energy (CERE), Faculty of Electrical Engineering & Technology, Universiti Malaysia Perlis, Pauh Putra Campus, 02600 Arau, Perlis, Malaysia
Aini Syahida Mohd Shaizad Department of Electrical Engineering, Faculty of Electrical Engineering & Technology, University of Malaysia Perlis, 02600 Arau, Perlis, Malaysia
Tunku Muhammad Nizar Tunku Mansur Department of Electrical Engineering, Faculty of Electrical Engineering & Technology, University of Malaysia Perlis, 02600 Arau, Perlis, Malaysia
Rosnazri Ali Department of Electrical Engineering, Faculty of Electrical Engineering & Technology, University of Malaysia Perlis, 02600 Arau, Perlis, Malaysia
Mohd Syahril Noor Shah Department of Electrical Engineering, Faculty of Electrical Engineering & Technology, University of Malaysia Perlis, 02600 Arau, Perlis, Malaysia
Mohd Zamri Hassan Department of Electrical Engineering, Faculty of Electrical Engineering & Technology, University of Malaysia Perlis, 02600 Arau, Perlis, Malaysia

DOI:

https://doi.org/10.37934/ctds.6.1.113

Keywords:

DSTATCOM, PQ Theory, harmonic current, conjugate gradient

Abstract

Power quality is important for distribution system as it may have negative impact for both the utility company and consumer if it has a power quality issues such as harmonic current distortion. This issue may result in a breakdown of equipment of both the consumer and the utility company. Thus, it is important to solve the power quality issue. Therefore, this research was conducted in order to overcome the power quality issues that is faced by the distribution system by adding DSTATCOM to the system as it can compensate harmonic current distortion at the point of common coupling (PCC). Conjugate gradient back-propagation neural network (GCBPNN) based PQ theory is chosen as the controller for this research as the effectiveness of DSTATCOM performance is depending on its controller. GCBPNN would reduce the time taken to compensate the harmonic. All the simulations for this research have achieved THD below 8% after adding DSTATCOM where the simulation that added GCBPNN achieved the lowest THD at 5.35%. These simulations are performed by using MATLAB/Simulink.