Digital Twin for Structural Health Monitoring (SHM)

In the latest of a series of Middlesex-Vietnam collaborations, MDX has received a New Institutional Links grant, under the Newton Programme Vietnam partnership. The grant is for a telecommunications research project in partnership with the University of Transport and Communications (UTC) in Hanoi. The Fund, provided by the UK Department of Business, Energy and Industrial Strategy (BEIS) and delivered by the British Council (for further information, please visit www.newtonfund.ac.uk) as part of the UK’s Official Development Assistance provision, supports activities that contribute to poverty reduction, welfare improvement or the progression of sustainable development in partner countries. The MDX-UTC initiative is to develop a low-cost Digital Twin model for early detection of structural damage to vital infrastructure such as bridges and dams that would be difficult to spot through monitoring of the physical structure alone.

The project, which begins on February 11 and runs for two years, will help improve quality of life for people in Vietnam and raise Middlesex’s research profile. It follows a Newton Fund-backed workshop in Hanoi in October involving British and Vietnamese academics, on the application of 5G technology to forecast accidents and manage traffic congestion, as part of the British Council-managed Researcher Links programme. The Principal Investigator is MDX’s Professor Huan Nguyen. The UTC team is led by Associate Professor Dr Thanh Bui, and the other partner is Stevenage-based technology firm Viavi Solutions, led by Dr Duc To and Dr Li-Ke Huang.

Summary

Service loads, environmental and accidental actions and natural hazards may cause damage to civil, lifeline infrastructures (lifelines are structures that are important or critical for a community to function, such as roadways, powerlines, bridges, dams, and port facilities). In developed countries, a civil engineering project often has a built-in structural health monitoring (SHM) system in the construction stage to ensure operational safety and to possibly reduce the cost related to life cycle management. However, in Vietnam, the management and the condition assessment of existing strategic infrastructures are rather neglected. This is because the annual budget allocation is prioritised to the rapid construction/development of new infrastructures rather than maintenance and monitoring. Recent structural collapses and major defects have caused great casualties, difficulties and concerns among local communities and general public. This proposal therefore aims specifically at developing an efficient but low-cost methodology that can help detect early the structural damages either by permanent or periodic monitoring. During the course of the project, Middlesex University (MDX) and the University of Transportation and Communications (UTC) will develop a low-cost structural health monitoring system by using vibration based field testing with Vietnamese conditions. Then, we will develop a twin digital model (so-called digital twin) to simulate what happens in a real lifeline structure (so-called physical twin). The aim is that the digital twin will have full function of many response outputs compared to limited sensory data as is the case for the physical structures. In this way, the monitoring of a real lifelines can be analysed by combining with the data from both models. This process will help reduce instrumentation in real structures and make the structural health monitoring low-cost and affordable in Vietnam.

Aim/objectives

The project aims to conduct vibration measurements that facilitate the development of a digital twin model in order to implement damage assessment for structural health monitoring (SHM) of important infrastructures (bridges and/or dams) in Vietnam. Specific objectives are:

  • Collecting existing databases and conducting new measurements for the long term vibration of real structures, i.e physical twin
  • Based on the collected data, a digital twin model will be developed based on big data analytics and machine learning (ML) algorithms
  • Statistical/data model-based approach is then used to connect the two models that can be used for damage detection and assessment of infrastructures
  • Validating the correlation between two models using both existing and new measurement data
  • Setting up datasets of structural data and tools for long-term damage assessment and permanent monitoring of important infrastructures in Vietnam

Work Packages

WP1 – Data collection and long-term vibration measurements

WP2 – Development of the digital twin model

WP3 – Physical based model and statistical data detection approach

WP4 – Testing and trial