Model reference adaptive controllers with improved performance for applications in LCL-filtered grid-connected converters

Abstract

This work presents a control strategy combining a direct Model Reference Adaptive Control (MRAC) with a robust state feedback for current control of grid-connected converters with LCL filters operating under grid impedance variations. The MRAC is based on a gradient law, providing properly current reference tracking and voltage disturbance rejection. The inclusion of a state feedback inner loop provides robust active damping of the filter resonance for an entire range of grid impedances. A systematic control design procedure is proposed, and the combination of both control actions in a multi-loop approach does not add significant computational complexity. The proposal allows to increase the MRAC gains adaptation dynamics, providing suitable responses under grid current reference variations and under typical grid voltage disturbances, including grid faults. Performance indexes as the grid-currents total harmonic distortion and the mean squared error are also improved. In addition, robust stability analysis based on Lyapunov functions are carried out for the inner control loop and for the outer control loop, considering the grid impedance uncertain and with unmodeled dynamics. The strategy is implemented in a commercial digital signal processor, and the compliance of the grid currents with the IEEE 1547 Std. is verified with controller hardware-in-the-loop and experimental results in a 5.4 kW prototype.