The selection of optimum screens for standalone screen (SAS) applications has historically been based on experimental data, rules of thumb, or correlations. Recent sand-retention tests conducted in various laboratories offer empirical screen-selection criteria on the basis of different sand-size-distribution parameters. Unfortunately, these experiments have their own limitations. They provide substantially different results, depending on how the tests are conducted and interpreted, leading to significant differences in the recommended screen type and screen-opening size for any given sand sample. To resolve these inconsistencies and to understand the physics of the problem better, this paper presents 3D numerical simulations to evaluate the performance of wire-wrapped sand screens and ultimately to develop systematic screen-selection criteria.

In this paper, a new method is presented to estimate the mass and size distribution of the solids produced through wire-wrap screens. The method uses the entire particle size distribution of the formation sand and is validated with experimental and numerical data. The new method allows us to evaluate the performance of different screens without running expensive and sometimes inconclusive experiments, enhances our understanding of screen performance, and helps to design sand screens better to meet performance criteria under a wide variety of conditions.

We first present results from 3D, discrete-element computer simulations of sand screens placed in contact with granular sandpacks of approximately 100,000 particles. The numerical model computes the mass and the size distribution of the solids produced. The effect of the most important parameters, such as friction coefficient, fluid viscosity, pressure gradient, and ratio of screen-opening size to sand size, on the mechanism of bridge formation and amount of sand produced is studied using both monodispersed and polydispersed systems.

The results have helped resolve some key questions about the physics of sand bridge formation. Numerous simulations are conducted to replicate the experimental conditions over a wide range of screen-opening/sand-size ratios for wire-wrap screens. Good agreement is observed between the laboratory experiments and the simulations.