Indirect Model Predictive Control of Grid Connected Modular Multilevel Converter

This paper deals with indirect model predictive control of grid connected modular multilevel converter (M
²
C). Indirect model predictive control is a simple and flexible control technique. To minimize cost function, it requires (N+1) switching vectors only while the direct model predictive control requires 22N switching vectors. Thus, the indirect model predictive control is less computational. Power mode control is designed to achieve predefined active and reactive power demands and to generate reference output currents. Performance of the control algorithm is demonstrated by step varying active and reactive power demands. Mathematical model of M
²
C is presented which is further used to implement indirect model predictive control. Finally, summary and simulation results for M
²
C are presented.