Abstract– This paper proposes an Adaptive Model Predictive Control (AMPC) framework to enhance the operation of grid-forming (GFM) converters in grid-connected applications. Virtual Oscillator Control (VOC) has emerged as a promising approach for GFM operation; however, these converters must achieve seamless transitions during grid connection and disconnection without introducing voltage spikes or inrush currents. By utilizing the multi-objective optimization capability of AMPC, a unified cost function is formulated for all operating modes, which simplifies the control structure and enhances transient performance. Building on this foundation and adjusting the weighting of different objectives in response to system conditions, the method achieves smooth mode transitions, stability, and accurate reference tracking. Simulation results verify the capability of the proposed method to operate effectively under different operational modes.
