Note: This content is accessible to all versions of every browser. However, this browser does not seem to support current Web standards, preventing the display of our site's design details.


Model-Mismatch in Model Predictive Building Control: Performance Sensitivity and Model Adaptations


R. Franz

Automatic Control Laboratory, ETH Zürich, Semester Thesis, FS 13 (10249)

In this report we investigate the effects of model-mismatch in model predictive building control and propose methods for adapting the MPC model. In the first step we observe performance sensitivity with regard to constraint violations and energy consumption. Furthermore we propose ideas for mitigating negative effects of model-mismatch for model predictive building control. We briefly explain thermal resistance capacity networks for the usage of modeling buildings and derive a mathematical formulation of a building model. Based on that, various factors which may cause model-mismatch are presented and we explain how they affect our previously defined building model. Then an introduction to MPC is given on the basis of a general description of the MPC problem which we are using for our simulations. We derive a nominal MPC problem and in addition we propose ways of modifying the nominal MPC problem in order to improve controller performance in presence of model-mismatch. We simulate the behavior of the nominal as well as the modified MPC controllers with and without different kinds of model-mismatch. The results of the simulations are presented and discussed. Finally, we summarize our findings and present ideas for future work.


Type of Publication:

(13)Semester/Bachelor Thesis

G. Schildbach

File Download:

Request a copy of this publication.
(Uses JavaScript)
% Autogenerated BibTeX entry
@PhdThesis { Xxx:2013:IFA_4629
Permanent link