Digital Twin as a Decision-Making Support Tool for Resilience of Urban’s Infrastructure under Extreme Climatic Events
MoRE2020 Fellow Kamyab Zandi, outgoing mobility from Chalmers University of Technology to Stanford University, USA
Transport infrastructures are the backbone of our society due to that we heavily rely on the uninterrupted availability of the road network, and bridges are the bottleneck of urban/rural mobility network. Extreme climatic events and climate change are important threats to the reliability and safety of the bridge infrastructures. This project will establish an innovative remedial action in which a Digital Twin of transport infrastructure is leveraged as a decision making support tool to minimize the impact of extreme events on seamless transport operation, increase the capacity of the road network for mobility of people and freight, protect the users of the transport network and provide optimal information to the operators. The Digital Twin brings all the data and models together providing powerful analysis and insight to decision making processes.
Västra Götaland’s transport infrastructure includes many large bridges, from Svinesund Bridge in the north to Älvsborg Bridge in the south, as well as the prospect of new landmark bridges such as Hisingsbron. Moreover, the bridge infrastructures in this region are not only exposed to extremely harsh climate – the harshest in Sweden both due to corrosive sea environment and freeze-thaw cycles – they are also vulnerable against weather-related flood events caused by heavy rains, flooded rivers and high sea levels. Therefore, this project contributes to sustainable urban development and the growth of Västra Götaland as a leading knowledge region. Moreover, the project will help Västra Götaland to take on the challenge of climate adaptation and to create a resilient urban region by developing next-generation decision making tools for its key infrastructures.
Collaborating end-users: the City of Gothenburg, the City of Stockholm
Summary of Project Results
Transport infrastructures are the backbone of our society due to the heavy rely on the uninterupted availability of the road network. Reliable condition assessment and maintenance of our aging transport infrastructures are essential to maintain their integrity and serviceability. Moreover, extreme climatic events and climate change are important threats to the reliability and safety of the road network. This has led to a growing demand for more accurate and reliable condition assessment - data collection, data interpretation and performance prediction - processes to ensure ever more resilient road transportation. This repository of data and models is integral to the framwork of Digital Twin for transport infrastructures.
A Digital Twin of an infrastructure is a living digital simulation that brings all the data and models together, and updates itself from multiple sources to represent its physical counterpart. The Digital Twin, maintained throughout the life cycle of an asset and easily accessible at any time, provides the infrastructure owner/users with an early insight into potential risk to mobility induced by climatic events, heavy vehicle load and even aging of a transport infrastructure.
In this project, we initiated a strategic cooperation between Chalmers University of Technology, Stanford University, as well as the City of Stockholm and the City o Gothenburg as the collaborating end-users. The complementary expertise of Structures and Composites Laboratory at Stanford in "collecting & analyzing data from structures" together with the expertise of Concrete Structures researh group at Chalmers in "developing models for the assessment of structures" were leveraged to develop tools that enable creation of Digital Twins of infrastructures. In this context, the collaboration has specifically resulted in:
- Data-driven models for quantification of cracks: An innovative data-driven model was developed and successfully tested. An important challenge was to create a large training database of cracks in concrete. Thus, much effort was put to create and annotate a 3D point cloud database of cracks. The data format, data quantity and data qulity are of unique features of our database. The models and the results were presented in a conference paper.
- Physics-based model incorporating condition assessment data: By building upon our earlier method of "discretizing tension softening curves", the method was tested in two frameworks: smeared crack approach and discrete crack approach, which were presented in a conference paper. Furthermore, a method was developed to generate FE mesh from 3D point cloud data in an automated procedure; this method was presented in another conference paper.
- Demonstration of Digital Twin concept in real scale: we successfully demonstrated our crack detection tool and Digital Twin framework in real-scale in IWSHM 2019 conference.
The outcome of the project was disseminated in 4 conference articles. Furthermore, two journal papers will be published to describe the theoretical development that our joint work has led to. Several authors from Stanford and Chalmers have contributed to disseminate the outcome of the collaboration. Last but not least, the collaboration has led to Chalmers/Stanford initiative "Digital Twins Lab - http://www.digitaltwinslab.com" through which we will follow up on our collaboration, disseminate the outcomes, seek for further funding and make alliances with new strategic partners.