An Improved Model Predictive Control Strategy for Virtual Synchronous Generator with Adaptive Parameters

Virtual Synchronous Generator (VSG) can simulate the characteristics of synchronous generator to provide inertia and damping for the system. For systems using an NPC three-level converter structure, traditional control methods require complex dual-loop control and internal PI parameter tuning, as well as more sensors. In addition, when island microgrid is switched by high power load, fixed parameter VSG cannot guarantee frequency regulation performance. To solve these problems, this paper proposes an NPC three-level VSG model predictive control strategy with parameter adaptation. The method is simple in operation and reduces the use of DC side current sensor. The angular velocity deviation and the change-rate of angular velocity are combined with the adaptive control strategy, and the fastest tracking differentiator is used to extract the angular velocity rate of change without setting subsection functions and thresholds, which effectively avoids frequent parameter fluctuations and improves the frequency stability of the microgrid when power changes. The effectiveness of the proposed strategy is verified by simulation.