Simple two-stage weighting factor design for finite control set model predictive control of modular multilevel converters

Due to discrete nature of power converters, finite control set model predictive control (FCS-MPC) is considered as an attractive choice for these systems. Since this control technique provides advantages like improved dynamic performance and inclusion of several control objectives in a single cost function, it is especially appropriate for topologies like modular multilevel converters (MMCs), where multiple control goals, that is, output and circulating currents control, and cells' capacitor voltage balancing should be satisfied. Despite these advantages, huge computational burden and weighting factor selection are two serious obstacles. In this work, a simple two-stage weighting factor selection scheme is suggested for FCS-MPC applied to MMCs. At the first stage, some offline tests are conducted for different weighting factors; and a pre-optimal value is selected, based on minimizing the total harmonic distortion (THD) of output current and root mean square (RMS) value of the circulating current. However, since optimal performance at different operating conditions is achieved by different weighting factors, the pre-optimal value will be updated at the second stage by a simple online algorithm. Moreover, a technique is suggested for effectively decreasing the computational burden. Finally, by conducting several simulation and experimental tests, the satisfactory performance of proposed controller is validated, compared to fixed weighting factor FCS-MPC.