A preliminary experimental study on sand production under true triaxial stress conditions
Sand production prediction is becoming a regular study in reservoirs with weak or unconsolidated sands. Three main approaches for sanding prediction are analytical, numerical and experimental methods. Laboratory experiments have proven to provide more realistic results, with these being used to understand sanding mechanisms and validate analytical and numerical methods. A large number of experimental studies have been carried out by researchers worldwide—most of which have been performed on cylindrical-shape samples under uniaxial (i.e. σ1 ≠ 0, σ2 = σ3 = 0) or triaxial (i.e. σ1 ≠ 0, σ2 = σ3 ≠ 0) stress conditions. In general, a sanding experiment under true-triaxial stresses (i.e. σ1 ≠ σ2 ≠ σ3 ≠ 0) is more realistic in simulating downhole conditions. This stress condition can be simulated in the laboratory on a cubic sample. The first part of this paper provides a comprehensive but brief literature review on past sanding laboratory experiments. This will be followed by the introducition of a unique true-triaxial stress cell (TTSC) which was modified and used for sanding simulations in the laboratory. The applied modifications will be illustrated and the test procedure described. The sample preparation for testing synthetic samples will be explained and some preliminary results obtained will be presented.