Data-driven modelling has emerged as a strong alternative to conventional modelling methods in various domains. It also brings a huge impact on the geotechnical and pushes researchers to re-think the potential of this new technology in engineering practice. Increasing research works have unveiled its versatility in modelling soil properties, behaviours, risk assessment, engineering construction and design. Meanwhile, this novel technology has also been incurring scepticism such as poor generalization ability and applicability in the engineering domain.
This special collection aims to provide a collection on the recent advances in data-driven modelling that could potentially support surrogate modelling, mechanical modelling, solutions to boundary value problems, the discovery of new physics, and provide reliable engineering solutions to address broad and unsolved problems in geoengineering, such as:
- Unified modelling of soil properties adapts to different sites;
- A new-generation data-driven constitutive modelling method;
- Intelligent and low-carbon construction materials design method;
- Data technologies with digital twins;
- Data-driven based uncertainty quantification method and reliability-based design;
- Intelligent construction and design in geoengineering;
- Machine learning algorithms enhanced numerical solver for computational geomechanics;
- Data-driven assisted discovery of new physics in geotechnical engineering.
In particular, we seek contributions that advance the state-of-the-art data-driven approach for geomechanics and geotechnical engineering in both big, sparse data regimes, and physics constraints.
