Wireless Sensor Networks for Building Information Modeling

Dimitrios Kontaxis, G.v. Tsoulos, Georgia Athanasiadou, Georgios B. GiannakisTelecom 2022 · DOI

TL;DR

This paper details the deployment and architecture of Wireless Sensor Networks (WSNs) integrated with Building Information Modeling (BIM) for energy-efficient building renovations. By collecting real-time environmental and energy consumption data from pilot sites, the research demonstrates how WSNs can enrich BIM with dynamic information, enabling more accurate occupant behavior modeling and improved energy performance simulations.

What Problem Does This Paper Address?

The construction industry is increasingly embracing "digitalization" through Building Information Modeling (BIM). However, current BIM applications often focus on static building information, limiting their effectiveness in optimizing energy-driven renovations for existing residences. To truly enhance energy efficiency, BIM needs to incorporate dynamic data streams related to indoor/outdoor conditions and energy consumption. This requires robust Wireless Sensor Networks (WSNs) capable of capturing and transmitting this real-time data to cloud-based renovation tools. The challenge lies in designing, deploying, and integrating these WSNs effectively within the BIM framework, considering practical field requirements and ensuring interoperability.

Methods

The paper focuses on the practical implementation and architecture of WSNs within the context of the BIMERR project. Key methodological aspects include:

  • WSN Technology Selection: The Z-Wave protocol was chosen for its suitability for indoor environments, low power consumption, and mesh networking capabilities, which allow for robust communication even in the presence of obstacles.
  • Network Topology Design: The WSNs were designed with a star-of-stars topology, where individual apartments formed local star networks connected to a central gateway, which then communicated with the BIMERR middleware via the internet.
  • Deployment Strategy: The paper details the placement of sensors (temperature, humidity, CO2, motion, door/window, power meters, etc.) within pilot apartments across three European countries (Poland, Spain, Greece), considering factors like optimal gateway location for signal strength and accessibility.
  • Data Acquisition and Processing: Real-time sensor data is collected by the WSNs and transmitted to a BIMERR middleware. This middleware handles data management, routing, and storage, making it available for further analysis.
  • Occupant Behavior Modeling: A key application of the collected data is training occupant behavior (OB) models using machine learning algorithms. This involves processing sensor data to understand occupant actions and preferences, which are then used to improve energy performance simulations.

Datasets

The paper does not introduce new benchmark datasets. Instead, it utilizes real-world data collected from WSN installations in pilot buildings across three European countries as part of the BIMERR project. These datasets include:

  • Environmental Data: Room air temperature, humidity, CO2 levels, outdoor temperature, and relative humidity.
  • Energy Consumption Data: Power consumption of individual appliances (e.g., air conditioning units) and total apartment energy consumption.
  • Building State Data: Status of doors and windows (open/closed), and HVAC system status (on/off, set-point temperature, operation mode).

This real-world data was collected over specific periods in the Greek, Spanish, and Polish pilot sites, serving as the basis for demonstrating the WSN's capabilities and training OB models.

Key Insights and Ideas

  • Bridging the Static-Dynamic Gap in BIM: The core insight is that integrating real-time sensor data from WSNs is crucial for moving BIM beyond static representations to dynamic, actionable insights for building energy management and renovation.
  • Practicality of Z-Wave for Residential WSNs: The paper demonstrates the effectiveness of Z-Wave technology for indoor residential WSN deployments, highlighting its mesh networking capabilities and suitability for the project's requirements.
  • Occupant Behavior as a Key Uncertainty: The accuracy of energy performance simulations is significantly impacted by occupant behavior. WSN-collected data provides the necessary granularity to model this behavior more accurately.
  • WSN Data for Predictive Modeling: The collected sensor data streams are directly used to train machine learning models that predict occupant actions and preferences, leading to more realistic energy consumption forecasts.
  • Interoperability is Key: The BIMERR project emphasizes an interoperability framework that allows seamless data exchange between WSNs, middleware, and BIM tools, crucial for a holistic renovation process.

Key Results

  • Successful WSN Deployment: The paper details the successful deployment of WSNs in pilot apartments across three countries, demonstrating the feasibility of collecting diverse real-time data.
  • Data Flow Validation: The collected sensor data streams were shown to flow seamlessly through the BIMERR middleware, enabling their use in subsequent analysis and modeling.
  • Occupant Behavior Model Training: The WSN data was successfully used to train occupant behavior models, specifically for thermal comfort and HVAC system adjustments, as demonstrated by the Greek pilot site's trained model.
  • Improved Energy Performance Estimation: By incorporating data-driven occupant behavior models, the accuracy of building energy performance simulations can be significantly increased, leading to better-informed renovation decisions.
  • Cost-Effective Solutions: The paper provides budget estimations for the WSN deployments, indicating that such systems can be implemented within reasonable project costs.

Limitations and Future Work

  • Limitations:
    • The accuracy of energy performance estimations is still highly dependent on the level of detail in the input data, with occupant behavior being a major source of uncertainty.
  • The Greek pilot site's initial data collection was impacted by the COVID-19 pandemic, leading to a period of building vacancy and biased evaluation of trained algorithms.
  • While Z-Wave was chosen, its data transfer rate (100 Kbps) is lower than some alternatives like ZigBee, though it was deemed sufficient for the project's needs.
  • Future Work:
    • Optimal WSN Deployment with BIM: Exploring how BIM technology can be used to automatically generate and visualize optimal WSN deployment solutions for smart buildings.
  • ML for Full Lifecycle BIM: Applying machine learning to BIM across the entire project lifecycle, from design to operations, to gain deeper insights and support decision-making.
  • Further Exploration of OB Modeling: Investigating a wider range of machine learning algorithms and modeling approaches for occupant behavior.
  • Standardization: Promoting the use of standardized data formats like obXML for occupant behavior data to enhance interoperability.

Summary updated

Abstract

Building Information Modeling (BIM) is a critical element for the “digitalization” of the construction industry and can be exploited for energy-driven renovation procedures of existing residences. Advancing beyond a BIM with data-capturing capabilities that are limited to building static information only requires sensor data streams related to indoor/outdoor ambient conditions, as well as to energy-consumption parameters of the residences. The data streams require the deployment of robust Wireless Sensor Networks (WSNs) that are able to capture and transmit real-time data to appropriate cloud-based renovation toolkits. The technology and topology of such networks are addressed herein. The paper sets the lines for similar installations that are required by the construction industry for collecting dynamic data, since it is based on the outcome of real-world WSN installations in pilot sites in three European countries, carried out in the context of a major collaborative BIM research project. An application example of the WSN data is also provided in the context of training occupant behavior models in order to demonstrate the use of the measured data.
BibTeX
@inproceedings{kontaxis2022wireless,
  author = {Dimitrios Kontaxis and G.V. Tsoulos and Georgia Athanasiadou and Georgios B. Giannakis},
  title = {{Wireless Sensor Networks for Building Information Modeling}},
  booktitle = {Telecom},
  year = {2022},
  doi = {10.3390/telecom3010007},
}
Figure