What is it about?
Model Predictive Control (MPC) is now the most popular advanced feedback control technology, in particular for processes which are multivariable and under constraints. The popular nonlinear models are in the form of state-space models. The necessary and key requirement is that MPC provides offset-free control, i.e. with zero errors in steady states. In the paper new MPC algorithms with this property are proposed, which provide simpler design, simpler control structure and competitive control quality when compared to the existing MPC design alternatives.
Featured Image
Why is it important?
Model Predictive Control (MPC) is now the most popular advanced feedback control technology. In practice, the processes are always under the influence of disturbances which are external or internal (modeling errors). The necessary and key requirement for every feedback control algorithm is that it provides offset-free control, i.e. with zero errors in steady states, under the influence of disturbances. MPC algorithms with this property are proposed in the paper, for nonlinear state-space models. This algorithms are simpler in design and lead to simpler control structures than the existing ones.
Perspectives
There was still in the recent years a lack of a clear understanding how the deterministic disturbances (which include modeling errors) can be effectively treated in the MPC algorithms with state-space process models. The paper concentrates on this problem, for the defined class of asymptotically constant deterministic disturbances - providing new solutions and their comparison with the existing alternatives.
Piotr Tatjewski
Politechnika Warszawska
Read the Original
This page is a summary of: Offset-free nonlinear Model Predictive Control with state-space process models, Archives of Control Sciences, December 2017, De Gruyter,
DOI: 10.1515/acsc-2017-0035.
You can read the full text:
Contributors
The following have contributed to this page
