International Journal on Engineering Applications (I.R.E.A.), Vol. 7, N. 6 ISSN 2281-2881 November 2019 Copyright © 2019 Praise Worthy Prize S.r.l. - All rights reserved https://doi.org/10.15866/irea.v7i6.17180 188 188 Power Quality Improvement with Mitigation of Harmonic Distortion in Three-phase System Based on Shunt Active Power Filter Mechanism Sakdawut Boontua, Teerawut Savangboon, Arckarakit Chaithanakulwat Abstract – In response to rising business cost and global environmental concerns, this research proposes a mechanism to improve power quality by reducing total harmonic distortion (THD) in three-phase three-wire system using shunt active power filter (SAPF). Three control algorithms are utilized to achieve a reduction in THD: DC-link capacitor voltage, hysteresis current, and harmonic extraction algorithms. The proposed SAPF mechanism is ideal for small-scale businesses and residential projects keen on energy conservation and natural resources preservation. In the study, simulations were carried out using MATLAB/Simulink. To verify, a prototype was fabricated and experiments were undertaken. The simulation and experimental results are in good agreement. The proposed SAPF mechanism could reduce THD of the system from 41.47% to 15.07%. Copyright © 2019 Praise Worthy Prize S.r.l. - All rights reserved. Keywords: Harmonic Current, Active Power Filter, Total Harmonic Distortion, Non-Linear Load, PI Controller Nomenclature APF Active Power Filter ADALINE Adaptive Linear Neuron DSTATCOM Static Synchronous Compensator PLL Phase Locked Loop P&O Perturbation and Observation PVG Photovoltaic Generator PCC Point of Common Coupling SAPF Shunt Active Power Filter THD Harmonic Distortion i s Source current i L Load current i com Command current i ref Reference current i feed Injection current i c Compensation current P fun Fundamental power P ref Reference power P har Harmonic power I s(max) Source current maximum i s(min) Source current minimum i s(loss) Source current loss v s Source voltage I. Introduction With advancement in electronic equipment and electrical appliances comes non-linear loads in the electrical system as a result current interruptions or harmonics. The harmonic or “pollution” in the electrical system shortens the life of electronic devices and contributes to loss of power in the internal electrical appliances [1], unnecessarily driving up the utility bills. Thus three harmonic harmonization schemes have been implemented to improve power quality (i.e., mitigate the harmonic distortion) and increase energy efficiency: passive filter circuits, active filter circuits, and high bridge power filter circuits. In [2], [3], the power quality improvement for large-scale industrial users was realized using static synchronous compensator (DSTATCOM) [17]. However, the technology suffers from high investment and fabrication costs due to its large scale. In [4], [5], an artificial neural network-based shunt active power filter (SAPF) was developed in accordance with IEEE standard and implemented in non-linear load micro-grid. In [6], [7], passive and active power filter flicks based on neural learning algorithms were comparatively experimented to improve the power quality of renewable-energy micro-grid single-phase and three-phase systems. The results revealed that the passive filters were bulkier and less cost-effective. To overcome the challenges, the passive filters were replaced with active power filters, which could effectively reduce total harmonic distortion (THD) [8], [9], [18]. In [10], the shunt active filter system (SAF) was implemented in photovoltaic systems to reduce harmonics and compensate for reactive power. Fuzzy logic was used to generate the reference current to control power (P) and reactive power (Q). The SAF is suitable for use in small manufacturing plants and residential buildings. In [11], parallel power filters (SAPF), together with unified adaptive linear neuron (ADALINE) algorithms, were implemented to efficiently improve power quality. In [12],