The petroleum industry spends millions of dollars each year to prevent and overcome sand control problems and their prediction has therefore become essential to minimize the adverse effects of sand production. This paper reviews existing experimental and modelling studies of the prediction of sand production. In experimental studies, the trapdoor system has been commonly used. The early experimental set-ups had a simple trapdoor configuration with a fluid inlet port to inject a single fluid to conduct qualitative tests. Other facilities were added over time, including facilities to conduct qualitative and quantitative tests, inject radial, axial and multiphase fluids and conduct tests in uniaxial and tri-axial stress environments. Experimental findings on sand production can be categorized into four different effects: the flowing fluid characteristics effect, the material characteristics effect, the loading effect, and the outlet size effect. However, the application of laboratory findings to field conditions has become doubtful due to the various assumptions used in laboratory studies, including the scaling down of the reservoir, the use of sieved river sands and/or synthetic samples, and the use of limited combinations of two-phase flows.
In model development, some basic mechanisms have been recognized as responsible for sand production: shear and tensile failure, the critical pressure gradient and drawdown pressure, and erosion. Tensile failure occurs when the effective minimum principle stress applying on the rock mass reaches its tensile strength, and shear failure occurs when the applied stress on some planes in the wellbore area is higher than their bearing capacity. Reduction of reservoir pressure causes the effective hydrostatic stresses on it to be increased, breaking its pore structure, and the rock formation can be eroded internally and externally, producing sand by dynamic seepage drag forces. To date, both continuum and discrete element method (DEM) approaches have been considered in sand production models. Most approaches have been developed considering the hydro-mechanical or shear failure mechanisms and some have combined them.
- Petroleum industry
- Predictive models
- Sand production