Data-driven Two-Step Identification of Building Thermal Characteristics: A Case Study of Office Building
Ziqing Wei,Fukang Ren,Yikang Zhu,Bao Yue,Yunxiao Ding,Chunyuan Zheng,Bin Li,Xiaoqiang Zhai
DOI: https://doi.org/10.1016/j.apenergy.2022.119949
IF: 11.2
2022-01-01
Applied Energy
Abstract:Thermal characteristics of building affect the energy consumption of air conditioning systems directly. Reverse grey box model is widely used for the identification of thermal characteristics of building. Prior studies have focused on predictive performance, but the reasonableness of the identified results is usually neglected. In addition, as an important characteristic, the air exchange rate is generally predetermined to reduce the complexity of the model because it always deviates from the design value during operation. For the purpose of overcoming the above problems, a two-step identification process based on resistance-capacity model is proposed in this paper. Three critical thermal characteristics, namely, lumped heat transfer coefficient, air exchange rate, and zone air heat capacity are identified by means of least squares method and analytical solution at different steps. The identification results of the three characteristics are 1519.71 W/K, 1242.39 m(3)/h, and 1513.56 kJ/K, with the error of 21.18%, 10.86% and 3.31%, respectively. The thermal characteristics in this paper are identified rationally while showing similar accuracy to the results from traditional resistance-capacity model. The proposed approach can be used to assess the reasonableness of the thermal characteristics of building for efficiency operation.