A Novel Adaptive Virtual Inertia Control-Based Adaptive Neuro-Fuzzy to Enhance Frequency Stability of a Microgrid with Seamless Transition

Abstract

A frequency stability improvement of a microgrid with low-inertia constant during severe disturbances remains a challenge using a virtual inertia control to compensate the inertia power to the microgrid, using a battery energy storage system. However, the fixed virtual inertia constant and the low-order model of the battery energy storage system utilized in virtual inertia control can affect the stability behaviors of microgrid and virtual inertia control design. Hence, for solving these issues, this paper proposes a novel adaptive virtual inertia control utilizing an adaptive neuro-fuzzy inference system based on a training data design in order to adjust the controller gain that applies for the direct and the quadrature axis model of the battery energy storage system. In order to investigate the proposed method and to verify the performance, the various case studies are carried out under the high, the low levels existing of renewable energy, and the parameter uncertainties in a microgrid. The simulated results with the proposed scheme are compared with the virtual inertia based-linear quadratic Gaussian synthesis and without the virtual inertia control. The comparative results reveal that the proposed method enhances the frequency stability during the disturbances, reduces the switching transition time from the islanded to the grid-connected mode of the microgrid, seamlessly, with superior performance in comparison with the other methods, satisfying a frequency grid code in Thailand.