Model Predictive Controller With Reduced Complexity for Grid-Tied Multilevel Inverters

Abstract

Finite control set model predictive control (FCSMPC) has become popular for grid interactive applications due to its simplicity and ease of implementation. Since the implementation of FCSMPC requires all possible switching states of the inverter, the computational burden is an important issue, which needs to be addressed while applying FCSMPC to the control of multilevel inverters. This letter proposes a novel “S” factor scheme for the FCSMPC implementation, which reduces complexity while controlling grid current and balancing the capacitor voltages in neutral point clamped three-level inverters. The proposed scheme is very easy to implement and can be extended to any higher level inverter without any additional effort. In this method, at every sampling instant, redundant switching states are chosen suitably and the predicted reference inverter voltage (PRIV) vector is derived from predicted grid voltage, predicted reference current, and the present value of the actual current. Subsequently, “S” factor scheme is applied to select the switching state, which is close to the PRIV vector. The selected switching state is then applied at the same sampling instant. The proposed scheme is experimentally compared with a recent FCSMPC algorithm to validate the performance.