Robust Optimized Current Controller Based on a Two-Step Procedure for Grid-Connected Converters

Proportional-resonant controllers are largely used for grid-connected converters with LCL filters to ensure sinusoidal reference tracking and harmonic disturbance rejection. However, the design of such controllers becomes more difficult when it is necessary to ensure stability and suitable performance for converters operating under uncertain grid conditions, from stiff to weaker, and also when the number of resonants increases. In this more challenging scenario, available design procedures may lead to a poor tradeoff between transient and steady-state performances. To overcome this problem, this paper proposes an off-line automated control design procedure to obtain multiple robust optimized proportional-resonant controllers. The proposed procedure consists of two steps, where a particle swarm algorithm is used to obtain: i) the gains of an inner loop ensuring optimal active damping for the LCL filter resonance; ii) the gains of multiple proportional-resonant controllers by minimizing a tracking error index, taking into account uncertain grid impedance and control signal limits. The robust stability of the closed-loop system with the control gains tuned by the metaheuristics is certificated through a parameter-dependent Lyapunov function, constructed based on linear matrix inequalities. Several results from control Hardware-in-the-loop are presented for a case study, revealing that the proposal provides better performance in terms of reference tracking and disturbance rejection, when compared to similar control strategies from the literature. Experimental results from a prototype connected to the utility grid attest the practical feasibility of the proposed procedure.