Enhancing Microgrid Small-Signal Stability and Reactive Power Sharing Using ANFIS-Tuned Virtual Inductances

Abstract

Microgrid as the main building block for future smart grids is prone to instability originated from converter-based distributed generations (DG). Herein, we first analyze the small-signal stability of an inverter-interfaced microgrid. Afterwards, a novel adaptive network fuzzy inference system (ANFIS)-based optimization method is introduced which aims at online tuning of virtual inductances (VI) in the islanded microgrids. The data for ANFIS training is drawn by particle swarm optimization (PSO) algorithm and the proposed objective function. A total of 140 load scenarios are considered to provide optimal VI in each load condition and generate optimal data for ANFIS training. This process minimizes reactive power mismatches and improves microgrid stability in different load levels. The simultaneous application of PSO algorithm and ANFIS training facilitates the objectives pursued by current study. Finally, the trained ANFIS networks are installed in the converter control. The adaptive performance of ANFIS controllers makes the converters responses independent from load change location and value. The effectiveness of the proposed control methodology is verified using simulations studies.