Parameter identification framework of thermal network model for ventilated heating floor

Fenggang Wang,Jiwei Guo,Yiqiang Jiang,Cheng Sun
DOI: https://doi.org/10.1016/j.enbuild.2024.114138
IF: 7.201
2024-04-11
Energy and Buildings
Abstract:Due to system complexity and the inherent disturbances of building environment, identification of key parameters of thermal model of building with air-based thermally activated heating floor is challenging. In this study, a two-stage framework for identifying unknown parameters in the thermal network model of building with ventilated heating floor is proposed. Besides, two estimation methods are compared, including least square method and Bayesian inference, and the impact of input data quantity and parameter bound is also analyzed. Results show that floor ventilation can significantly reduce temperature fluctuation and energy consumption due to the increased space heating rate. Compared to least square method, Bayesian inference fully consider uncertainties, and can improve the accuracy. In addition, it is found that 9 days' data is accurate enough for parameter identification of this heating system.
energy & fuels,construction & building technology,engineering, civil
What problem does this paper attempt to address?