Multiscale modelling of shelly carbonate sands for foundation design of offshore structures (MuMShell)

Lead Research Organisation: City, University of London
Department Name: Sch of Engineering and Mathematical Sci

Abstract

Carbonate soils cover over 40% of the world's seabed, where offshore structures, pipelines, artificial islands and other marine structures are founded. For the most part, carbonate soils are of biogenic origin comprising skeleton bodies and shells of small organisms, the shelly carbonate sands. These soils are a complex and poorly understood material as evidenced by a number of accidents reported during platform installation in the 80s. As a consequence, shelly sands have been placed into a niche classification of "problematic soils" in most design guides. While failures are now relatively rare, conservative methods and high factors of safety are commonly used. Understanding the behaviour of shelly carbonate sand is critical for the design of foundations for offshore structures. In particular, understanding the physical phenomena taking place at the microscale has the potential to spur the development of robust computational methods to be integrated into novel or existing design approaches.

Image-based geomechanics is a fascinating research field that has the potential to transform the way soils are investigated and modeled. The ability to follow deformation at the microscale has helped to answer fundamental questions about the soil behaviour observed at the macro-scale. The proposed research uses 4D synchrotron x-ray imaging and post analysis to investigate the kinematics and the strain maps of a shelly carbonate sand under compression. The outcomes will contribute scientific understanding on the multiscale behaviour of shelly carbonate sands. This will form the basis to develop fabric-informed constitutive models to better predict the soil response, thus improving design practices for foundations of offshore structures. The ambition of this project is to contribute towards safer, less conservative and more sustainable ground structures and reduce the financial risks associated with unforeseen ground response during construction of offshore foundations. This multiscale methodology and image algorithms here developed, will be valuable to the broad granular media community to simulate mechanical processes in additive manufacturing, mining, food and pharmaceutical industries.

Planned Impact

This research project has been designed to have an impact beyond the academic environment. It is anticipated that the short and mid-term beneficiaries of this research project will be:

Marine and offshore engineering industry, including:
- Design engineers of offshore foundations will have access to advanced multiscale computational tools for shelly carbonate sands to incorporate in their design approaches. This will improve design practices, thus contributing to enhance business revenue and the innovative capacity of the UK offshore industry, noting that UK engineers are engaged with foundation design around the world.
- Contractors of offshore structures will benefit from better design practices for shelly sands, which will avoid additional cost associated with over-designed foundations.
- Companies working on site investigation technology will benefit from further investment in R&D on soils with complex mechanical behaviour, stimulated by this research
- Consultancy companies in the UK and abroad will benefit from hiring graduate students with a better knowledge of soils of biogenic origin as a result of the teaching material and textbooks complemented with the outcomes from this project

Granular media and materials processing industry, which will benefit from:
- A better simulation of processes such as compaction and grain damage/fracture of granular products following further R&D actions motivated by the multiscale methodology and algorithms available from this research project
- A new generation of scientists engaged with multiscale approaches for innovative industry-oriented research, which will contribute with advanced processing and handling solutions for granular materials with a clear advantage to the UK and rapid emergent economies.

Longer term socio-economic impact of this research project is also anticipated and it is envisaged that this might include:
- Safer, cost effective and more sustainable construction practices for offshore engineering with a direct impact on i) the UK economy and ii) on the quality of life of communities living in regions rich in shelly carbonate deposits.
- The lower cost of products, such as food and medication, resulting from the optimisation of the processing and handling of granular materials from the perspective of time and energy saving.
- An increased number of graduate engineers, encouraged to study STEM subject areas by high-tech and exciting research as this project, which will contribute with practical solutions to the 21st century challenges of sustainability, housing and an ageing population.
 
Description The advanced in-house codes developed in this project were instrumental to quantify for the first time in the literature the microstructure of shelly carbonate sands. The results show that the techniques developed successfully overcomes major challenges posed by the diversity and complexity of the shapes associated with the bioclastic nature of shelly sands. In particular, the key advantages of the segmentation algorithm used to identify the individual grains are: (a) its iterative nature that enables an image to become progressively segmented, and (b) the use of truly adaptive parameters that are determined at the local rather than global level. Given the ability of the technique to deal with extreme grain morphologies, it can be readily used to segment other granular materials. New shape parameters were proposed, which shown to capture well the variety of grain shapes and to provide more intuitive and meaningful metrics. In particular, the newly proposed angularity parameter based on grain surface curvatures was found to provide a good description of the corners and sharp edges of shelly grains. The image-based approach developed enables more accurate measurements of intra-granular void ratio and grain size distribution, when compared with invasive experimental methods.

We have used a DVC-reconstruction method to quantify grain kinematics in shelly carbonate sands for sands. This technique has the advantage of overcoming limitation related to the complex morphologies and/or proneness to grain damage, for which, for which IDtracking methods can be of limited application. The rich image-based experimental data from this study will be instrumental for the development and validation of advanced discrete numerical approaches to model the behaviour of shelly carbonate sands.

We have carried out preliminary work to develop a numerical tool to model granular material with complex morphologies, such as shelly carbonate sands, which takes into account the detailed representation of grain shape and internal voids.
Exploitation Route The developed codes can be used by researchers in granular media and materials processing. The novel insights gained into the microstructure of shelly sands can be used in textbooks. Design engineers of offshore foundations have now access to a more accurate characterisation of carbonate sands, which will help them to tailor site investigation to the unique properties of the soil, which, as demonstrated in this project, differ significantly from more common silica sands.
The numerical tool developed here can be further extended to obtain a more accurate modelling of granular materials.
Sectors Agriculture, Food and Drink,Construction,Education,Energy,Environment,Manufacturing, including Industrial Biotechology,Pharmaceuticals and Medical Biotechnology

 
Title Adaptive watershed segmentation technique 
Description This is a tool to identify individual objects in 3D images (image segmentation) that was developed with the aim of overcoming limitations of current available methods. The basic principle consists of performing a series of iterations that enable the segmentation to become progressively more refined, while largely allowing for under-segmentation in each iteration. In this method the concept of 'region' is used, which is defined as a cloud of connected solid voxels in the binary image that is not in contact with any other solid voxels. Each iteration consists of two main operations: (a) identifying all the regions in the binary image, and (b) for each region, modifying the inversted distance map and performing watershed segmentation accordingly. 
Type Of Material Improvements to research infrastructure 
Year Produced 2017 
Provided To Others? Yes  
Impact This will enable improving both characterisation and modelling of granular materials. 
URL http://openaccess.city.ac.uk/17899/
 
Description Micro-mechanics of hollow grained hemipelagic calcareous sediments 
Organisation University of Massachusetts Dartmouth
Country United States 
Sector Academic/University 
PI Contribution The colleagues at University of Massachusetts Dartmouth are using the computer codes developed in this project to investigate sands from Australia
Collaborator Contribution The colleagues at University of Massachusetts Dartmouth have proposed a new direction to this reserach using neural networks
Impact Thsi collaboration is multi-disciplinary, we are working with colleagues from computer science and gelogists, this is ongoing research and there are no outputs yet.
Start Year 2018
 
Description GM3 2017 - Micromechanics from Micro to Macro, City University of London 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact There were 50 attendees including academics and industry participants, for the most part from the UK but also from Hong Kong and the US.
Year(s) Of Engagement Activity 2017