Improved Power Sharing Strategy for Parallel Connected Inverters in Standalone Micro-grid

Abstract

The primary goal of integrating alternative energy systems such as solar and wind turbines into the power grid using power electronic devices is to meet the growing energy demands. Connecting inverters in parallel effectively enhance power capacity, reliability, and overall system efficiency. However, an uneven power distribution among the inverters is a significant limitation in these parallel connected inverters (PCI). This study focuses on a distributed generation (DG) unit comprising a solar photovoltaic system (SPV) and a battery energy storage system (BESS) connected to voltage source inverters (VSI) 1 and 2. The proposed approach aims to achieve uniform load/power distribution among the inverters with power management, maintaining a constant DC link voltage despite variations in solar irradiation and temperature. Additionally, the strategy targets the reduction of total harmonic distortion (THD) in the load current. The conventional droop control method is restricted in its ability to achieve accurate and uniform power distribution during load changes. An enhanced P-f and Q-E-based droop control technique (EDCT) and a fuzzy logic controller (FLC) have been proposed to address these challenges. Performance analysis using MATLAB/Simulink was conducted with two test cases, and a comparative assessment was carried out with a conventional controller, such as a Proportional Integral Controller (PIC) and Sliding Mode Controller (SMC), to showcase the effectiveness of the developed control technique. The proposed control method ensures equal sharing of active power of 2.5 and 4kW, 2.5 and 7.5kW and reactive powers of 2.5 and 4Vars, 2.5 and 7.5Vars among the inverters for the two cases even during the load change at t=0.5seconds. The settling time for DC link voltage and THD is effectively diminished to 0.03 and 0.25 seconds and 1.18% and 2.87% for the two test case studies of the proposed method, which are much lower than those of existing methods available in the literature.