Design of a Prediction-Accuracy-Enhanced Continuous-Time MPC for Disturbed Systems via a Disturbance Observer

Abstract

This paper addresses the prediction accuracy enhancement problem in the context of continuous-time model predictive control (MPC) by using a disturbance observer. The disturbance estimation is introduced in the output prediction to correct the errors caused by disturbances and uncertainties, which eventually leads to the desired offset-free tracking performance. Due to such a different design philosophy, the computational burden of the proposed method will be significantly reduced as compared with the existing offset-free MPCs for disturbed systems. Another promising feature of the proposed method lies in its nominal performance recovery property. The workability of the proposed method is validated by its application to a dc-dc buck converter system. Index Terms-Continuous-time model predictive control (MPC), disturbance observer (DOB), offset-free tracking, prediction accuracy enhancement.

Experimental Validation of a Robust Continuous Nonlinear Model Predictive Control Based Grid-Interlinked Photovoltaic Inverter

“…On the other hand, it can be shown that the resultant optimal control does not depend on the values of q 1 and q 2 [23]. Therefore, one can choose q 1 = 1 and q 2 = 1 although both outputs are with different magnitudes.…”

“…Nowadays, the closed-form CNMPC becomes more mature and there exists a steadily increasing interest in real time implementation of this technique to nonlinear systems with fast dynamics such as electrical drives and power converters [23]. In [24], a composite controller consisting of CNMPC and a disturbance observer is implemented in real-time to a permanent magnet synchronous motor, and improved performance is reported.…”

Advances in Disturbance/Uncertainty Estimation and Attenuation [Guest editors' introduction]

“…The work [5], authored by Yang, et al, addresses the predication accuracy enhancement problem using continuous-time model predictive control and a disturbance observer. The disturbance estimate is introduced in the output prediction, which leads to the desired offset-free tracking performance.…”

N-Step MPC for Systems With Persistent Bounded Disturbances Under SCP

“…Over the past few decades, due to its practical insights, the model predictive control (MPC) problem has attracted considerable research interest in systems with bounded disturbances, which gives rise to robust MPC (RMPC), see [4], [10], [17], [35] for discrete-time systems and [21], [26]- [28] for continuous-time systems. Up to date, RMPC strategy has been extensively applied into various engineering areas such as chemical process, power systems, DC motors and mobile robots [8], [10]- [14], [31], [36].…”