Search

To achieve carbon neutrality quickly and deal with urban greenhouse effect, it is necessary to study the energy consumption of urban buildings. Energy consumption prediction is an important step in planning and managing energy use in buildings. At present, few studies have comprehensively considered the relationship between geometric features, functional types, location and energy use of buildings in urban built-up areas. In this paper, a spatial data-driven framework is introduced for the analysis and assessment of building energy consumption. The framework is grounded in the building centroid and is combined with single photon radar and point of interest data through the utilization of the nearest neighbor classification algorithm. Two megacities, Beijing and Shanghai, are taken as examples to illustrate the method and to generate the regional building energy use database. The results show that: a) building physical feature (geometric form, functional type) have a significant impact on energy consumption, and energy consumption is correlated with the economic level of the region; b) Energy use in urban built-up areas has obvious spatial agglomeration characteristics, and the siphon effect is obvious in region with high energy consumption. The contribution of the framework is to provide insights into the feasibility of employing multiple spatial data fusion for the analysis of energy consumption within urban built-up areas. This can be employed to address energy planning concerns and help promote sustainable environmental practices in urban areas worldwide. • This paper attempts to provide a feasible framework for analyzing urban building energy use by utilizing real-time urban semantic perception information and building features. • The proposed framework have comprehensively considered the relationship between geometric features, functional types, location and energy use of buildings. • The proposed methodology can obtain regional energy consumption stocks and produce spatial distribution results, thus improving the interpretability of the data. • The results show that energy use in urban built-up areas has obvious spatial agglomeration characteristics, and the siphoning effect is obvious in regions with high energy consumption. [ABSTRACT FROM AUTHOR]

Group travel behaviour widely exists in cities, but has not been well investigated by researchers. To fill this gap, this paper develops a co-existence-based methodological framework to systematically explore the spatiotemporal characteristics of group travel behaviour. We apply our framework to a case study of Shanghai, China, using a one-month tranche of metro smart card data. Results show that most travellers perform a small number of group trips, together with a small number of co-travellers. They usually travel in a dyad or triad group and form far more small social communities than large ones. Group travel behaviour is distinctly different from individual travel in terms of both time and space: group travellers are more likely to travel during weekends, on holidays, and in the afternoons and evenings. They also prefer to perform group behaviour near stations located in the city centre or the centres of new towns in suburban areas, and close to attractions and public facilities. The analysis we present has various potential applications such as improving the management of public events and supporting the design of group ticket policy. [ABSTRACT FROM AUTHOR]