Digital twin of the world’s first 3D printed stainless steel bridge

Introduction

The 3D printed bridge is a cutting-edge architectural, engineering and data science experiment, pushing the boundaries of conventional infrastructure manufacturing processes and incorporating a state-of-the-art sensor network to monitor structural behaviour. This project is an interdisciplinary research endeavour for exploration and innovation in multiple scientific fields, drawing together experts from data sciences and statistics, engineering, materials science, industrial design, computer science and philosophy of technology, to collaborate on a project with revolutionary potential.

Printed by Dutch 3D printing company MX3D and developed as a collaboration between a number of industrial and academic partners, in addition to its unique construction, the 12 metre-long stainless steel bridge has been equipped with a sensor network designed and installed by Arup, Imperial College London, Autodesk, University of Twente and Force Technology, in collaboration with the Turing’s data-centric engineering programme, sponsored by the Lloyd’s Register Foundation.

Once the bridge is placed across the Oudezijds Achterburgwal canal in central Amsterdam it will immediately become a living laboratory. As part of the fabric of the local urban infrastructure, everyone that crosses over the bridge will generate data that will inform the development of a digital twin of the bridge to measure, monitor and analyse the performance of the structure.

Explaining the science

Developing a digital twin

The Alan Turing Institute’s involvement with the bridge began with the engagement of researchers on the data-centric engineering programme applying a data-centric approach to test the 3D printed stainless steel used to construct the bridge to ensure its safety, using statistical techniques in conjunction with materials science.

The novel 3D printing technique pioneered by MX3D represents a radical shift from conventional manufacturing, by making it possible to build new innovative large structures. Placing material only where it is essential makes for much stronger, more economical and unique structures. However, there are questions to be addressed about the steel’s material properties, due to its inherent variability and how to guarantee safety standards for its manufacture and continued monitoring.

Researchers on the data-centric engineering programme performed detailed experiments to probe the material properties of 3D printed stainless steel and to analyse the sensor data gathered using novel statistical methods, leading to improved understanding of this advanced material. The load tests showed the bridge was capable to hold a 19.5 ton load, which was well over its ultimate design load. These experimental tests provided critical evidence to secure a permit from the City of Amsterdam for bridge placement.

The data from the sensors is being fed into an intelligent digital twin, which will learn from live data, including strain, displacement, vibration data and environmental factors such as air quality and temperature collected from the bridge while in operation. The digital twin builds on mathematical models and abstractions already widely used in structural engineering and blends them with recent statistical and machine learning techniques. The digital twin of the bridge will be a living computer model which is continuously learning to imitate the physical bridge with growing accuracy as the data comes in.

The performance and behaviour of the physical bridge will be tested against its digital twin, which will provide valuable insights for the design and certification of future 3D printed structures. It will also enable the current 3D bridge to be modified to suit any required changes, ensuring it is safe and secure for pedestrians.

Beyond the digital twin of the bridge, the overarching vision of the data-centric engineering programme is to develop the essential mathematical foundations and software engineering needed for intelligent digital twins for all kinds of engineering applications.

To enable these digital twin studies, the Turing is developing a data access platform that will integrate with Autodesk’s software to enable researchers to access the large quantities of sensor data stored on cloud computing servers. The Turing is hosting the bridge data for the full two-year period covered by the bridge’s current operating permit and has conducted a thorough ethics review of the project to ensure that the scientific goals of the project do not compromise the public's privacy. Using a custom data platform, the Turing supports researchers who require access to the sensor data stored in its secure cloud.

Digital twin of 3d printed bridge — Predicted vertical deflections, under typical pedestrian loading, from the finite element digital twin model developed by the Steel Structures research group in the Department of Civil and Environmental Engineering, Imperial College London.

Interacting with the community

The project has engaged with the University of Twente’s BRIDE (BRIdging Data in the built Environment) project to use the bridge, its design and instrumentation to gather insight into how the city’s population engages with the bridge. This will subsequently inform how designers, technologists, and citizens can utilise rapid urban manufacturing and IoT technologies for designing urban space. By developing an understanding of the interactions between people, place, activity and technology that contribute to a sense of ‘cityness’, the project will analyse how smart technologies such as those used to create and monitor the bridge might contribute to designing 'cityness'. Ultimately this aims to promote a feeling of ownership of the public space by the community.

The 'mini-bridge'

Researchers have also developed a smaller-scale model of the 3D printed bridge, complete with sensor network and digital twin. This ‘mini-bridge’ acts as a demonstrator that can be toured to public venues to demonstrate the concept of a digital twin.

Project aims

With the improved knowledge from the statistical techniques from this project, the engineers and designers working on the bridge and future 3D printed steel structures, will have a richer understanding of how such structures can and will function.
Through multidisciplinary and collaborative working the project intends to demonstrate how taking a data-centric approach to design and construction of infrastructure – combined with ongoing monitoring – can richly inform the engineering processes, develop new techniques, and improve long-term safety.
The innovative statistical techniques applied to testing the advanced materials used to construct the bridge aim to contribute to the development of new standards for a new type of infrastructure with new materials. This has the potential to open up the construction industries to pioneering ways of working.
By involving a team of social science researchers, the project will inform the future development of urban infrastructure through gaining greater understanding of how a community interacts with smart technologies and innovative construction.