Predictive controller Using modal series for DC/AC converters

The real processes encountered in industry, have nonlinear behavior. For the sake of simplicity, generally a linear model around the operation point is developed in the controller design procedure and then the linear controller is applied to the real process. In this paper, an NMPC controller is designed for a DC/AC converter in state space. First, an MPC controller is designed based on a linearized state space model of the converter. Then using Modal Series the linear controller is modified such that the effects of nonlinear dynamics, which are not considered in the linear model, are compensated. Finally, the proposed method is compared with the EDMC approach. Results show that while the proposed method is comparable with the EDMC in the sense of system performance, its less computational capacity makes it a more attractive option for fast processes.