Performance Improvement of Predictive Direct Power Control of a Unified Power Quality Conditioner

Abstract

This paper presents an analysis and comparison of classical Direct Power Control (DPC) and Predictive Direct Power Control (P-DPC) strategies applied to the parallel side of a Unified Power Quality Conditioner (UPQC). Classical DPC is known for its simplicity and direct control of active and reactive power but has limitations in harmonic suppression and dynamic response under varying grid conditions (Chaoui et al., 2008 ; Gui et al., 2020 ; Zhao et al., 2019). The P-DPC strategy overcomes these challenges by predicting system behavior and optimizing control actions via a cost function, resulting in faster dynamic responses, enhanced harmonic mitigation, and improved power quality. On the series side, an adaptive hysteresis control method is combined with a synchronous reference frame identification technique. This dynamically adjusts the hysteresis band to ensure precise voltage compensation and reduce switching losses. Additionally, a fuzzy logic controller regulates the DC-link voltage, maintaining stable operation during transient and steady-state conditions. Simulation results demonstrate that P-DPC significantly outperforms classical DPC in harmonic suppression. Moreover, adaptive hysteresis control on the series active filter further improves voltage compensation, reducing total harmonic distortion (THD) compared to fixed-band hysteresis. The fuzzy logic controller offers robust DC-link voltage regulation, enhancing stability during disturbances. The integration of P-DPC on the shunt converter, adaptive hysteresis control on the series one, and fuzzy logic regulation of the DC-link voltage collectively demonstrate superior performance in harmonic mitigation, reactive power compensation, and voltage stabilization. These findings highlight the potential of predictive control techniques combined with intelligent controllers to achieve optimal power quality in modern electrical systems.