This paper proposes a novel secondary voltage control using nonlinear
multivariables robust adaptive control based on multiple models with unmodeled
dynamics for microgrids. The proposed secondary control scheme consists of a
linear robust adaptive controller, a neural network based nonlinear adaptive
controller, and a switching mechanism. The linear controller assures
boundedness of the input and output signals, and the nonlinear controller can
improve the tracking performance. By using a specially designed switching
scheme between the two controllers, it is demonstrated that the stability and
performance can be achieved simultaneously. By leveraging a data-driven
real-time identification, the proposed method does not require the information
of primary control and microgrid models, thus exhibiting well robustness,
generalizability, and disturbance-resistance.