COMPARATIVE ANALYSIS OF MODERN DIGITAL FILTER USED IN SEIG-DSTATCOM BASED DG FOR POWER QUALITY ENHANCEMENT

Abstract

Localized renewable energy resources, essential for rural electrification and energy security, will be rapidly integrated to autonomous distributed generation networks worldwide. The three-phase SEIG is the most popular electromechanical generating unit for the isolated applications due to its rugged, brushless construction and low capital cost. However, autonomous SEIG microgrids are highly susceptible to the injection of high frequency discontinuous harmonic currents generated by non-linear loads. Such harmonics cause severe non sinusoidal voltage distortions at the Point of Common Coupling (PCC) risking to demagnetize generators, and violating the stringent 5.0% source current THD limit, as required by IEEE Std 519-2014. A shunt connected DSTATCOM is implemented with a new mathematical simplified digital control architecture to reduce such extreme power quality anomalies. The Synchronous Reference Frame Theory (SRFT) is the traditional basis for the active power conditioning. While mathematically correct, SRFT depends on extremely weak Phase Locked Loop (PLL) synchronization and Park’s coordinate transformations. For low-inertia microgrids, these non-linear tracking loops are sensitive to loss of synchronization due to short voltage sags. The discontinuous Infinite Impulse Response (IIR) low-pass filters generate large asymptotic settling delays and delay the dynamic compensation. For overcoming these systemic mathematical constraints, the simulation and exhaustive evaluation of the Adaptive Vectorial Filter Theory (AVFT) control algorithm is especially designed for standalone SEIG networks. The logic of the AVFT arises from the stationary three-phase reference frame. Real time in-phase voltage unit templates are obtained and the optimized discrete Moving Average Filter (MAF) in one fundamental time cycle is aggregated to completely isolate the fundamental load weight and avoids being a recursive algorithmic phase lag. In the MATLAB/Simulink environment, a rigid side-by-side comparative analysis was made. In steady state condition SRFT controller reduced baseline uncompensated THD of 27.26% to 2.68%. However, the AVFT algorithm was able to reduce the source current THD to a highly optimized value of 1.97% which was better for harmonic neutralization. The results validate the better computational efficiency, dynamic agility of the MAF based AVFT architecture, and steady-state compliance for stabilizing highly volatile autonomous distributed generation systems.