| Autor: |
Mats Vande Cavey, Lieve Helsen, Roel De Coninck |
| Rok vydání: |
2014 |
| Předmět: |
|
| Zdroj: |
Linköping Electronic Conference Proceedings. |
| ISSN: |
1650-3686 |
| DOI: |
10.3384/ecp140961295 |
| Popis: |
Optimal control using Modelica models has promising opportunities with the development of JModelica. A model predictive control framework for optimally controlling a floor heated building heated by a heat pump is proposed. The control inputs are applied to virtual building emulator model with a limited amount of measurements. State estimation is implemented using a moving horizon estimation to reinitialize the states of the controller model in every timestep. To use the moving horizon estimation, the implementation of Modelica equations is altered.The Modelica state equations are supplemented with a stochastic input to represent the process noise (model error). The state estimation significantly improves the output matching between emulator and controller model. The JModelica optimization framework proves to be satisfactory for the limited size virtual case. Future work will be able to build on this framework to handle different models and prediction error. ispartof: International Modelica Conference location:Lund, Sweden date:10 Mar - 12 Mar 2014 status: published |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
