Parameter Optimization of Universal Droop and Internal Model Controller for Multi Inverter-Fed DGs Based on Accurate Small-Signal Model

TLDR: In this paper, a parameter optimization of universal droop and internal model control (IMC) is proposed based on an accurate small-signal model for an inverter dominated microgrid.

Abstract: Microgrid comprises of several distributed generations (DGs), which are typically integrated through power electronic inverters. The existence of low inertial devices combined with the dynamic nature of the load challenges the stability of a microgrid and the effectiveness of the controller, mainly when operated in islanded mode. It is essential to optimize the parameters of the controller to enhance its efficacy under various operating conditions. In this paper, parameter optimization of universal droop and internal model control (IMC) is proposed based on an accurate small-signal model for an inverter dominated microgrid. In order to achieve robust control performance under different load conditions, a four-step approach is proposed: 1) an accurate small-signal model of a parallel multi-inverter system is prepared, which operates with the universal droop and internal model controller. The developed small-signal model is more accurate because it considers the dynamics of filter and phase-locked loop; 2) an investigation of critical control parameters of universal droop and internal model controller influencing the system stability is carried out, and their corresponding stability domain is identified through eigenvalue analysis; 3) particle swarm optimization (PSO) is used to optimize the critical parameters; and 4) the obtained result is validated under different load disturbances. Following the above approach, the time domain simulation is performed, which establishes that the proposed scheme improves the dynamic response of the DGs, counteracts the disturbances effectively and simultaneously improves the power-sharing. The proposed model is also compared with the well-established conventional PI-based droop controller, which demonstrates the efficacy of the proposed scheme.