Grid-connected converter is the interface between new energy and power grid. The grid current is often disturbed by periodic interference, which leads to the power quality degradation. In order to reduce the current harmonics, a compound control strategy based on linear active disturbance rejection and repetitive control is adopted in this thesis. Because the compound control strategy has many parameters to be set, their physical meaning is not clear too, and the setting parameter process is seriously complicated. A parameter optimized scheme based on particle swarm optimized algorithm is proposed in this thesis. In this scheme, three fitness functions are suggested which are about relative value of parameter, the rapidity and steady-state error of the system, as well as the phase margin and harmonic suppression characteristics of the system. They are designed to meet the multi-objective optimization for the relative value of parameters and the dynamic and static characteristics of the system. This scheme makes the parameter setting be simplified greatly, and the time of setting parameter lower. The simulation and experimental results show that the proposed scheme is correct and effective.