Urban Building Energy Modeling: A City-scale Distributed Simulation Approach

城市建築能耗建模:基於城市尺度的分佈式仿真方法

Student thesis: Doctoral Thesis

View graph of relations

Author(s)

Detail(s)

Awarding Institution
Supervisors/Advisors
Award date23 Jul 2024

Abstract

Urbanization poses a significant challenge in the 21st century. Currently, more than half of the global population resides in urban areas, and this percentage is projected to reach 68% by 2050. The increase in urban population has led to a substantial rise in residential energy consumption, alongside a surge in commercial energy use to meet the growing demand for services. Consequently, overall building energy consumption has witnessed a significant increase. Therefore, effectively managing energy use in urban buildings has become imperative. To achieve this goal, various methodologies and tools for urban building energy modeling have been developed. These models offer valuable insights into the energy demands of building stock, covering benchmarking analysis, scenario assessments, peak load evaluations, energy pattern analysis, and other specialized analyses.

Despite extensive research in the field of energy modeling, assessing urban energy remains complex due to three significant challenges. Firstly, urban building simulation involves various aspects such as geography, construction, materials, and HVAC (Heating, Ventilation, and Air Conditioning) systems, each of which is stored in its own unique data model. As a result, creating text-based simulation files for urban buildings from scratch is an intricate task which requires the integration and processing of cross-domain data models. Secondly, conventional simulation models rely on climate conditions provided by a limited number of weather stations, which do not accurately capture the microclimate variations caused by urban morphologies, natural conditions, and man-made structures. This limitation results in unrealistic and unreliable simulation outputs, further hindering effective decision-making for urban sustainability. Lastly, previous efforts have primarily focused on complex physical conditions within cities but have often encountered challenges such as intricate modeling and substantial computational loads.

To address these gaps, this dissertation proposes a system architecture for urban building energy distributed simulation. The first aspect involves designing ontologies using semantic network technology, grounded in the features of building energy simulation inputs, clearly defining the potential logical relationships between the inputs, and facilitating the generation of qualified simulation files. Additionally, the concept of UrbanPatch, which represents the microclimate perception domain of urban buildings, is introduced. By analyzing the building morphology and green spaces within each UrbanPatch, a microclimate tuning approach is proposed to localize weather conditions for buildings. Finally, a rapid simulation approach is created, which decomposes the city model into spatially correlated building blocks for distributed simulation. The proposed algorithm, known as distributed adjacency blocks (DABs), uses 2D footprints to construct 3D building groups and considers solar azimuth angles, altitude angles, and shading planes to simplify the simulation targets. Using multiple threads and abstracted inter-building boundary conditions, the energy dynamics of an entire city can be simulated in parallel.

The innovative system architecture for urban building energy distributed simulation proposed in this dissertation offers a novel solution that prompts researchers to reconsider the traditional bottom-up approach towards city-scale energy simulation. Centered around distributed building networks, this dissertation not only distributes the computational load across multiple computing components, enabling dynamic energy simulations for extensive metropolitan areas, but also accounts for the influence of microclimate on building energy consumption in the urban built environment, resulting in more precise and reliable simulation outcomes and enhancing the efficiency of city energy decision-making and management.

    Research areas

  • Urban building energy modeling, Microclimate, Urban energy dynamics, Inter-building effects, Urban morphology, Urban heat island, Ontology, Semantic web technology