2019 CSCE Annual Conference - Laval (Greater Montreal) Conference
Dr. Ayan Sadhu, Western University
Structural Health Monitoring (SHM) data provides rich and critical information for assessment of existing conditions of structures. Currently, the majority of structural inspection methods use printed checklists, and their interpretation is labor intensive subject to personal judgement and prone to error. One of the primary concerns with SHM is huge amount of raw and processed data that has to be effectively managed using appropriate tools. To reach the full potential of structural inspections, the data management process has to be automated. Building Information Modeling (BIM) is a powerful data management tool that can provide a systematic digital environment for SHM for several purposes such as procurement, summarization, sharing and recalling of data in future. Such a hybrid information modeling platform integrates the architectural, engineering and construction systems of a structure into one place. However, current BIM-based software rarely go beyond construction phase. BIM models are treated as static information sources that contain as-built data. Various opportunities to get accurate information about the state of the structures can be explored by representing the real-time building information and making the BIM based models dynamic.
The main objective of this paper is to take a step forward from static towards dynamic BIM by managing and representing the data of SHM systems in real-time using BIM. The real-time monitoring of sensor data is crucial for assessing the state of the structure including various physical (i.e., stiffness) and modal (i.e., frequency or damping) parameters under different traffic and weather conditions. The workflow developed in the study is used to connect the sensor data with the Industry Foundation Classes (IFC) based BIM model through “virtual sensors” that can facilitate the interoperability of BIM models. The data is fused into the IFC based BIM model and valuable SHM information such as maximum acceleration, sampling frequency, number of vehicles passing over, etc. are available for each sensor in the BIM model. Furthermore, Matlab is integrated with BIM model such that system identification can be performed using any selected measured data of a cloud network in real-time. A long-span bridge in Thunder Bay is utilized to demonstrate the proposed methodology and illustrate the benefits of automated SHM using long-term monitored data in a unified framework.