Journal of Building Performance Simulation

In an era characterized by the widespread adoption of Building Management Systems (BAS) and Internet of Things (IoT) devices, coupled with advancements in data storage and processing technologies, the data landscape within the realm of building infrastructure has grown increasingly diverse. Data is drawn from various sources and sensors, including Building Information Modeling (BIM), BAS, IoT sensors, and simulations, which collectively capture both the static and dynamic attributes of buildings. The former encompasses geographical, geometric, and thermal characteristics, such as location, building dimensions, and envelope materials, while the latter primarily originates from BAS or field sensors, providing insight into operational data trends.

Recent developments in data-driven methodologies have paved the way for precise digital world modeling, fostering intricate mappings between virtual models and physical systems within the framework of a Cyber-Physical System (CPS). Within the domain of building management, these data-driven algorithms have found applications in energy management, energy consumption/load forecasting, and fault detection and diagnosis.

The amalgamation of multi-source building data offers a wealth of multidimensional information to fuel data-driven digital models, enabling the capture of both spatial and temporal nuances in target buildings. Moreover, the cross-validation of data from diverse sources enhances reliability. Consequently, the fusion and utilization of such multi-source data becomes indispensable in digital modeling, bridging the divide between single-source data-driven simulations and real-world physical systems.

However, multi-source data is often characterized by disparate formats, naming conventions, and granularities, introducing complexity and protracted development timelines. The creation and deployment of algorithms frequently necessitate substantial manual effort, tailored to the idiosyncrasies of each building.

Efforts to address these challenges should include:

1. Exploration of multi-source data fusion techniques, such as the establishment of standardized Building Model Integration (BMI) structures.
2. The development of comprehensive multi-source data integration strategies for digital twin modeling.
3. The formulation of multi-source data integration approaches to enhance building performance simulations, encompassing energy consumption/load prediction, thermal modeling, indoor air quality modeling, and energy systems modeling.
4. The design of multi-source data integration strategies to bolster performance evaluation and fault detection.
5. The investigation of multi-source data integration methods for intelligent building energy management, encompassing optimization of supply-side system operations and demand-side management, among other areas of exploration.

These endeavors collectively constitute a promising avenue for research and innovation within the field of multi-source data integration for digital world modeling.

Instructions for AuthorsSubmit an Article