Naslov
Regression-Based Thermodynamic Model Identification of a Zone with a Closed-Access Air Conditioner

Autori
Hure, Nikola ; Vašak, Mario

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
The Proceedings of the 30th IEEE Mediterranean Conference on Control and Automation / - , 2022, 336-342

Skup
30th Mediterranean Conference on Control and Automation (MED 2022)

Mjesto i datum
Atena, Grčka, 28.06.2022. - 01.07.2022

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
A/C and zone model, data-driven model, identification, regression models, regularization, experimental validation

Sažetak
Smart buildings have a great potential in the energy regulation market. One of the levers that are used for the flexibility provision of buildings are the thermal comfort systems. This paper deals with a thermodynamic model identification for a comfort- regulated zone of a smart home with an installed air conditioner. The intended end-use of the model is model predictive control of comfort with electricity demand response over a collection of objects with such similar configuration. Typical commercial setups of closed-access air conditioners found in residential objects are considered, where there is no possibility of any data communication from the air conditioner and where the sensory equipment is quite limited due to an intended large-scale deployment. This drives the specific input-output model form where the air conditioner electricity consumption is the selected model input. The performance of different mathematical models for temperature prediction in the smart home in the heating season is analyzed and the results with quantitative measures are provided. The complete analysis is based on measurement data collected on a smart home experimental setup.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika, Temeljne tehničke znanosti

POVEZANOST RADA