This work focuses on the design of a nonlinear model predictive controller (NMPC) and an H_∞ robust controller for an MEA-based CO₂ capture process. The model used in the NMPC is a nonlinear, additive, autoregressive model with exogenous (NAARX) inputs. Uncertainties are unavoidable in chemical processes. Therefore, a robust controller is designed for the CO₂ capture process based on μ-synthesis with a DK-iteration algorithm. The effects of uncertainties due to measurement noise and model mismatches are evaluated for both the NMPC and robust controller. The system disturbances include a number of input and output disturbances such as the flue gas flowrate and composition and variable capture targets. This study shows that the tradeoff between the fast tracking performance of the NMPC and the superior robust performance of the robust controller must be considered while designing the control system for the CO₂ capture units.

此工作的重点是一种非线性模型预测控制器（NMPC）和设计ħ _∞为基于MEA-CO鲁棒控制器₂捕获过程。NMPC中使用的模型是具有外来（NAARX）输入的非线性，加性，自回归模型。在化学过程中不可避免地存在不确定性。因此，为CO ₂设计了一种坚固的控制器基于μ合成和DK迭代算法的捕获过程。对于NMPC和鲁棒控制器，都将评估由于测量噪声和模型不匹配而引起的不确定性的影响。系统扰动包括许多输入和输出扰动，例如烟气流量和成分以及可变捕获目标。这项研究表明，在设计用于CO ₂捕集单元的控制系统时，必须考虑NMPC的快速跟踪性能与鲁棒控制器的出色鲁棒性能之间的折衷。