Search

Urban building energy modeling (UBEM) is a promising tool to evaluate building energy consumption at high temporal-spatial resolution at urban scale. However, state-of-the-art UBEMs use identical building envelope thermal transmittance according to design specifications in model characterization, which neglects the uncertainty in U-values of the building stock. To evaluate the impacts of uncertainty in U-values on heating/cooling demand at the urban scale, this study proposes a physics-based model of building envelope thermal transmittance to generate the U-value distributions of building stocks. Diversified U-values were assigned to buildings according to the building age, while identical initial U-values in the design served as a baseline. The impact of uncertainty in building envelope U-values was assessed by comparing the heating/cooling demand simulated under different U-value distributions, taking 33,222 residential buildings in Beijing as a case study. The results showed that, considering uncertain physical factors, thermal transmittance followed a right-skewed distribution, which led to an increase in the heating/cooling demand of the building stock. The annual total heating/cooling demand increased by 26% and 13%, respectively. The diversity in the heating/cooling demand intensity of the building stock was enhanced when more causes of uncertainty were considered, which was more evident in the building stock with a small range of building age or heating demand intensity. Therefore, it is advisable to consider the uncertainty in the building envelope thermal transmittance in UBEM simulations for energy evaluation and planning at the district or urban scale. [ABSTRACT FROM AUTHOR]

[Display omitted] • Presented a novel co-simulation framework for ground-coupled building envelopes. • Developed an XGB-based regression model for heat responses over various buildings. • Derived the theoretical thermal process solutions for equivalent slabs. • Integrated the underground heat transfer model into the building simulation engines. • Validated the modeling accuracy and computational efficiency of the proposed method. Ground-coupled building envelopes contribute to up to 45% of total building heat loss. Dynamic simulation of the thermal process of the building-ground structure is essential for whole-year building energy simulations and evaluations. There exist various types of underground heat transfer models, while the adoption of dynamic underground heat transfer models in building energy simulation software is quite limited because of accuracy or computational efficiency issues. To fill this gap, this research proposes a novel co-simulation framework for dynamic underground heat transfer with building energy models. An XGB regression model was developed for the calculation of the heat response coefficient of building-ground structures, based on which an equivalent slab model is proposed and integrated with the general building energy models. The proposed co-simulation model was validated using the detailed 3-dimensional finite difference method. It can be concluded from the validation results that the simulation errors of the inner surface temperature of ground-coupled envelopes do not exceed 0.6℃, while the computational speed is improved by 4000–20000 times. The proposed model was integrated into the building performance simulation software DeST for the whole-year dynamic simulation of building-ground structures. [ABSTRACT FROM AUTHOR]