Model predictive power control with Kalman filter for grid‐tied inverter with parameter variation

Abstract

Model predictive power control is considered to be a promising control algorithm owing to its simple structure and fast dynamic response. This algorithm relies heavily on the predictive model and the parameter variation adds the difficulty to gain the accurate model, which affects the static performance of the grid‐tied inverter. Here, a model predictive power control with Kalman filter (MPPCKF) is proposed to address above issue. On the basis of the grid‐current predictive model, the mathematical model is gotten to estimate the AC‐filter inductance and a Kalman filter is introduced to suppress the noise interference of calculating inductance. Finally, a simulation and an experimental platform are established to demonstrate the validity of the proposed approach. The results exhibit that the MPPCKF can effectively suppress the grid‐current distortion and the active‐/reactive‐power ripple of grid‐tied inverter caused by the mismatched predictive power model.

A model predictive control of three‐phase grid‐connected current‐source inverter based on optimization theory

“…The gridconnected current-source inverters (CSIs) act as an interface between renewable energy and the power grid, which has a greater impact on the energy conversion system. Compared with voltage-source inverters (VSIs), CSIs have the advantages of boost capacity, reliable short-circuit protection capacity, and direct-current control capacity [1][2][3], which have been widely utilized in power conversion applications such as renewable power systems [4,5], electric drives [6,7], and active filters [8].…”