Mesoscopic modelling of erosion in particle-laden flow systems

Nearly 70% of hydrocarbons are produced from poorly consolidated sandstone reservoirs worldwide which are inevitably mixed with sands. Sand production - as small as 0.01w\% - leads to major erosion problems such as degradation of pipelines, down-hole tubing and production equipment including pumps and valves. The impinging particles (usually in the range 10-300 microns) act as miniature machine tools which cut into the surface material and generate material chips [10], causing excessive damage to the surface layer and reducing the effectiveness of corrosion inhibitors. The enhanced corrosive degradation will in turn accelerate erosion resulting in severe and rapid loss of surface metal (washout). A lack of real-time and local measurement options along length of the fluid-carrying pipelines makes prevention of catastrophic failure difficult.
The erosion mechanisms depend on the interplay of multiple factors such as particle collision rate, size, shape, impinging angles and velocities, and particle-surface material properties. A recent review of 23 different erosion models concludes that many of the aforementioned parameters are not adequately included in the models due to a lack of fundamental understanding of the underlying physical mechanisms. This has resulted in the development of excessively conservative models which are adversely affecting efficiency and manufacturing costs of oil/gas production lines and equipment. Therefore, we urgently need predictive capability to assess the service life of existing oil/gas production flow systems, and to optimise the design of new ones.
However, a step-change of our capability in predicting erosive impact of particles will only be possible if their dynamical behaviour can be captured both individually and collectively. The key research question is therefore to understand the dynamical role of impinging solid particles in the erosion process, enabling quantitative prediction of erosive impact of particles and mass removal from the surface. We will tackle this research challenge by developing a new validated multiscale computational tool.
We aim to transform the current ad hoc approach to modelling erosion by developing a totally new computational tool based on first principles to quantify erosive impact of particles on surfaces in flow systems. Therefore, we set out the following objectives:

- To correlate the flow turbulence and the rate of particle-surface collisions by extending the CFD-DEM approach in the regions of high particle concentration.
- To determine the erosive effect of impinging particles on surface, the peridynamics theory will be utilised to assess surface damage, which will be calibrated and validated experimentally.