E-PolyU501/24
Geotechnical Resilience through Intelligent Design

Hong Kong Principal Investigator: Prof Zhenyu Yin (The Hong Kong Polytechnic University)
European Principal Investigator: Dr Enrico Soranzo (Institute of Geotechnical Engineering BOKU)

The resilience of geoinfrastructure is increasingly imperilled by the impacts of climate change. For instance, within the EU, current damages amount to €0.8 billion annually, a figure projected to surge by 1500% by the century's end. Within Hong Kong, climate change-induced extreme rainfall has frequently occurred in recent years, leading to numerous flooding disasters, landslides, etc., inducing significant economic and environmental damage. Consequently, there is a pressing need for a substantial enhancement of geotechnical engineering methodologies and knowledge in anticipation of a world marked by heightened uncertainty.

Geotechnical engineering has long grappled with three inherent challenges (the 3T challenges): uncertainty stemming from incomplete knowledge, heterogeneity arising from diverse geomaterial compositions, geological processes and settings, and nonlinearity resulting from complex interactions. The compounding effects of climate change exacerbate these challenges, impeding traditional analytical and numerical approaches in accurately predicting geomaterial behaviour, crucial for designing resilient infrastructure, decision-making and conducting effective risk assessments. While machine learning (ML) has emerged as a promising tool across various disciplines, its application in geotechnical engineering encounters barriers due to the complexity of the field and the aforementioned inherent challenges, compounded by data limitations associated with climate change.

This proposed research initiative seeks to propel ML into the forefront of geotechnical engineering, with a vision to address critical challenges and revolutionise the field for the betterment of society. The overarching goals of the project align with the need to confront uncertainty, combat climate change through zero carbon emission strategies, address soil parameter heterogeneity, expedite finite element (FE) calculations, e.g., for reliability analyses, and enhance design efficiency to reduce material consumption, particularly in the context of concrete. By undertaking this multidimensional approach, the research aims not only to apply ML in geotechnical engineering but also to fundamentally transform the field, ushering in a new era of efficiency, sustainability and resilience. Through collaboration and innovation, we aspire to make ML an integral and indispensable tool for addressing the complex challenges faced by geotechnical practitioners in the 21st century.

European partners have been granted the Geotechnical Resilience through Intelligent Design in the MSCA Staff Exchanges program, enabling their researchers to visit PolyU and collaborate with us on this topic. Since PolyU is the partner outside of Europe, the EU cannot fund the HK team to visit European partners. Therefore, we are optimistic that this project will provide the necessary support to facilitate such visits in the future.

E-PolyU502/24
Geometric Finite Element Methods

Hong Kong Principal Investigator: Prof Buyang Li (The Hong Kong Polytechnic University)
European Principal Investigator: Dr Kaibo Hu (University of Edinburgh)

Tensor-valued PDEs on surfaces have significant theoretical value and applications in a wide range of fields, including nonlinear elasticity, general relativity, computer graphics, and differential geometry and topology. This proposal aims to develop stable, efficient, and high-accuracy numerical schemes for tensor-valued partial differential equations (PDEs) on manifolds by developing intrinsic tensor finite elements and advanced numerical analysis techniques for solving tensor-valued PDEs on manifolds, particularly intrinsic geometric flows, such as the Ricci flow, a fundamental tool in differential geometry and topology. The research aims to provide a rigorous theoretical analysis of the numerical approximation, including stability, convergence, and error estimates. The outcomes of this research are expected to advance the numerical analysis and practical applications of PDEs on manifolds.