Digital Twin for Infrastructure using Big Data and Smart Sensor Networks

Project: Digital Twin for Infrastructure using Big Data and Smart Sensor Networks

Funding: VinGroup – VinIF

Duration: 2022-2024

Main partners: University of Transport and Communication (Vietnam) and London Digital Twin Research Centre (Middlesex University London)

PI/Co-PI: Assos. Prof. Thanh T. Bui (UTC) and Prof Huan X. Nguyen (LDTRC)


During operation, structures may be damaged, reduced operational efficiency and life expectancy because of adverse impacts such as corrosion, overload, environmental loads, natural disasters, etc. The damage of these structures not only has severe impacts on costs but also public safety. Structural health monitoring (SHM) is an effective tool to ensure that structures can operate efficiently throughout the design lifecycle. Over the last decade, SHM has been commonly applied and achieved success around the world. However, because of the current tendency to monitor the health of the entire structure to increase the accuracy, instead of focusing solely on suspicious locations, the amount of collected data is extremely growing. This is posing challenges to the data acquisition and analysis techniques. For instance, the Vincent Thomas Bridge in San Pedro, California uses 26 sensors for SHM generating about 3 terabytes (TB) of data per year. The bridge monitoring project in the Russian Federation extracts about 7 gigabytes (GB) of data per day. More than 20 GB of data collected during an automated rail test in the city of Brockton, Massachusetts or the SHM of wind turbine blades in Belgium generated more than 300 GB of data in six months.

In this proposal, we will build on the success of our previous Newton Fund project to advance the developed digital twin model further to handle the big data obtained from smart sensors to monitor the structure health. Digital twins create living digital  models that update and change as their physical counterparts change. A digital twin continuously learns and updates itself from multiple sources to represent its near real-time status, working condition or position. In this project, we will focus specifically on the big structural data and the design of  smart sensors for SHM purpose (including the edge computing technique). Physical characteristics of the intact and damaged structure are identified, located, and quantified based on obtained data . ML will be also used to overcome the limitation of traditional approaches, employing the long-term data measured on structures, to develop the methodology that can evaluate and monitor the health of the infrastructures.


The specific objectives are

    • O1 – Data collection for digital twin model
    • A2 – Deploying ML and edge computing for digital twin model
    • A3 – Prognostic health monitoring