An accurate and fast model is essential for implementation of model predictive control (MPC) in building management systems to reduce the load on electric grid without sacrificing occupant comfort. In this paper, an open source emulation model (made available on GitHub) has been developed for a Danish university teaching building to be applied in the MPC toolchain BOBTEST within the IBPSA project 1. The model is calibrated and realistic occupancy is introduced based on data measured by meters installed in the building. Simulation and sensitivity analysis performed on controllers and electricity-consuming devices helped to improve the emulator implementation and make it BOBTEST-ready. Results indicate that (1) the model is physically realistic in terms of electric load and thermal comfort estimation and (2) a new control/actuation setup considerably improves simulation time.
|Titel||2020 6th IEEE International Energy Conference (ENERGYCon)|
|Publikationsdato||29. okt. 2020|
|Status||Udgivet - 29. okt. 2020|
|Begivenhed||6th IEEE International Energy Conference (ENERGYCon) 2020 - Gammarth, Tunesien|
Varighed: 28. sep. 2020 → 1. okt. 2020
|Konference||6th IEEE International Energy Conference (ENERGYCon) 2020|
|Periode||28/09/2020 → 01/10/2020|