Model predictive flexible power control for grid‐connected inverter under unbalanced network

Abstract

Model predictive power control (MPPC) is considered as a promising algorithm utilised in grid-connected inverter due to its fast dynamic response, simple control structure and multi-objective optimisation. However, under unbalanced network, the grid current with excessive distortion troubles the application of MPPC. Based on the conventional MPPC, this study proposes a model predictive flexible power control (MPFPC) to reduce the current total harmonic distortion and achieve three flexible targets, including the cancellation of the active- and reactive-power oscillation and the negative-sequence grid current. In this algorithm, the power references are divided into two parts, which are used as reference values for the instantaneous active power control (IAPC) and the instantaneous reactive power control (IRPC), respectively. The three flexible targets are achieved by regulating the ratio of the two parts of power reference. Meanwhile, the grid current of MPFPC is the sum of IAPC and IRPC; therefore, the grid current distortion can be suppressed effectively. Simulation and experimental platform constructed by a three-level inverter are established to validate the effectiveness of MPFPC, and the results exhibit that the MPFPC meets the above expectations.