PolyU project promotes multidimensional machine learning in geotechnical engineering supported by RGC via EU Co-funding Mechanism
image: Prof. Zhen-Yu Yin, Professor of the Department of Civil and Environment Engineering at PolyU.
Credit: © 2024 Research and Innovation Office, The Hong Kong Polytechnic University. All Rights Reserved.
Ensuring the resilience of geo-infrastructure is a critical global task amid the challenges brought about by climate change. A project led by The Hong Kong Polytechnic University (PolyU) is leveraging machine learning and artificial intelligence (AI) to advance geotechnical engineering, positioning these technologies as next-generation tools to tackle complex challenges. The research has been supported via the European Union (EU) — Hong Kong Research Cooperation Co-funding Mechanism by the Research Grants Council (RGC) 2024/25.
The joint project - “Geotechnical Resilience through Intelligent Design” - is led by Prof. Zhen-Yu YIN, Professor of the Department of Civil and Environment Engineering at PolyU, in collaboration with Dr Enrico SORANZO, Deputy Head of the Institute of Geotechnical Engineering at BOKU University in Austria, and has been awarded HK$497,600 for a duration of 48 months.
Geotechnical engineering faces inherent challenges due to the diverse compositions of geomaterials, complex geological processes and the nonlinear interactions among them. These challenges are further intensified by the impacts of climate change, with the attendant rise in the frequency of extreme weather events.
This project aims to train the next generation of machine learning-savvy geotechnical engineers and assemble an interdisciplinary, intersectoral team that connects industry and academia. Together, the team will develop transformative approaches to infrastructure design, including for tunnels, tailing dams, artificial slopes and embankments, as well as geohazard monitoring and prediction through innovative data-driven methods.
Key initiatives of the project include exploring advanced sensor technologies for on-site data acquisition, leveraging generative AI techniques to optimise design processes and implementing physics-informed neural networks to enhance the fidelity of geotechnical simulations by integrating physical principles into machine learning models.
The project additionally focuses on designing customised solutions for specific challenges and tailoring machine learning approaches to address geotechnical issues encountered in the construction and management of geo-infrastructures, such as tunnels and tailing dams. It also aims to improve the monitoring and mitigation of geohazards, such as landslides and earthquakes, thereby enhancing safety and efficiency in critical areas.
Prof. Yin said, “By employing a multidimensional approach, the study aims not only to apply machine learning in geotechnical engineering but also to fundamentally transform the field, ushering in a new era of efficiency, sustainability and resilience.”
The European Union Hong Kong Research Cooperation Co-funding Mechanism by the RGC aims to strengthen collaboration between European and Hong Kong research communities in areas of mutual interest in order to achieve world-class scientific results.