论文标题
基于谐波的模型预测控制
Harmonic based model predictive control for set-point tracking
论文作者
论文摘要
本文提出了一种新型的模型预测控制（MPC）公式，以进行设定点跟踪。基于终端成分的稳定预测控制器在小型到中度预测范围的参考变化时可能表现出稳定性和可行性问题。在用于跟踪公式的MPC中，通过添加人工平衡点作为新的决策变量来解决这些问题，从而提供了一个明显扩大的吸引力领域，并确保了任何参考变化的递归可行性。但是，如果预测范围不够大，它可能会遇到性能问题。本文提出了该公式的扩展，其中使用谐波人工参考代替平衡点。所提出的配方达到了更大的吸引力领域，当预测范围很小时，可以显着优于其他MPC公式。我们证明了所提出的控制器的渐近稳定性和递归可行性，并为其主要成分设计提供了指南。最后，我们通过对球和板系统的案例研究强调了它的优势。
This paper presents a novel model predictive control (MPC) formulation for set-point tracking. Stabilizing predictive controllers based on terminal ingredients may exhibit stability and feasibility issues in the event of a reference change for small to moderate prediction horizons. In the MPC for tracking formulation, these issues are solved by the addition of an artificial equilibrium point as a new decision variable, providing a significantly enlarged domain of attraction and guaranteeing recursive feasibility for any reference change. However, it may suffer from performance issues if the prediction horizon is not large enough. This paper presents an extension of this formulation where a harmonic artificial reference is used in place of the equilibrium point. The proposed formulation achieves even greater domains of attraction and can significantly outperform other MPC formulations when the prediction horizon is small. We prove the asymptotic stability and recursive feasibility of the proposed controller, as well as provide guidelines for the design of its main ingredients. Finally, we highlight its advantages with a case study of a ball and plate system.
