Seabed ploughing is currently used for the installation of pipelines and cable-laying in the oil/gas and telecom industries. Total global pipeline and control line Capex is expected to increase by 59.8% over the next five years compared to the period 2008-2012, with developments in Europe set to account for the largest share of this, with a clear long term shift towards SURF (subsea, umbilical, risers and flowlines) infrastructure, through the development of remote fields by the UK and Norway. With the anticipated increase in offshore energy installations, by 2050, activity in seabed ploughing will have increased considerably in the UK alone.
Despite this activity, and that expected in coming years, the means by which key parameters to design a ploughing scheme are determined (for pipelines or cables) is limited. With improved modelling we expect major savings could be made, and that is the key goal of this proposal. In the past 20 years, remarkable improvements have been realised in understanding the mechanical and hydraulic processes in many soil-structure interaction problems. For instance, in shallow soft ground tunnelling we are now able to predict the damage that will occur to buildings with much greater confidence. This has been achieved by advances in computational modelling, validated by experimental and field measurements.
Unlike many other soil-structure problems (e.g. tunnelling, slope stability) soil ploughing has not been subject to significant numerical modelling, probably because there are features of the problem which make life very difficult. A major barrier is that the problem is not strictly amenable to any 2D simplification, and in the cases of most interest to industry involves coupled mechanical and hydraulic behaviour. The seabed ploughing problem therefore remains one in which there is relatively little guidance on the key parameters which determine economic feasibility, i.e. tow force, speed and stability. Currently, the only tool available to contractors planning ploughing programmes is based on a semi-empirical approach and relies on soil parameters for which field data are limited. The goal of this project is to develop a numerical model of ploughing at Durham University, validated through experimental investigations at the University of Dundee and guided by a collaboration of experienced industry partners, which can replace this approach.
The key features required of a numerical model of ploughing are: (a) material nonlinearity, (b) geometric nonlinearity (i.e. finite deformation mechanics), (c) temporal discretisation and (d) spatial (3D) discretisation. Any mesh-based method (such as the finite-element method) will struggle due to mesh distortion as a result of (b) and will be computationally expensive due to the need for remeshing during a simulation. As indicated earlier we see this as a barrier that has held up development of numerical modelling of ploughing, largely because research has focussed on finite-element methods and has so far ignored alternatives developed in the computational mechanics community.
An alternative to mesh-based methods are “meshless” methods such as the Element-Free Galerkin method, however these remain computationally uncompetitive and not suitable here. This project will use a hybrid finite-element, meshless method, the material point method (MPM), that contains aspects of both finite-elements and meshless methods.
The objectives of the project are:
- to produce a high quality experimental data set at the University of Dundee of model ploughing at 1g scale, backed up by centrifuge testing;
- to develop a computational method at Durham University, the Material Point Method, to include coupled hydro-mechanical behaviour, material nonlinearity and advanced soil constitutive models;
- to verify the computational method against existing analytical solutions and/or published results;
- to validate the computational method against the experimental testing and field data;
- to use the model to inform design of seabed ploughing and plough share design;
- to disseminate the outputs and findings from 5 to the research community and industry.