An Improved Active Disturbance Rejection Model Predictive Power Control with Circulating Current Reduction for Grid-Connected Modular Multilevel Converter

In this paper, a new combined control strategy based on model predictive control and active disturbance rejection control has been proposed to control a modular multilevel converter used for interfacing a floating photovoltaic system into the power grid. The proposed control consists of a combination of the active disturbance rejection controller for DC link voltage regulation and an optimized model predictive control strategy for power flow control and circulating current reduction for modular multilevel converter. The great number of submodules in the MMC converter increases the complexity of the MPC control and the calculation time for choosing the voltage vector to be synthesized according to the required cost function. In this paper, an optimized method is proposed to minimize the time of calculation and application of the vector voltage reference, this method is based on a local optimization for each elementary hexagon. A study is carried out in order to illustrate the performances of the proposed control strategy. A multiple case study approach is adopted to show the superiority of the active disturbance rejection model predictive control strategy in terms of rapid convergence, elimination of overshoot, and improved robustness under various disturbances and operating conditions.