When higher than expected sand volumes were produced from Gas Field #1 after a 2nd well group was brought on line it was initially believed that the sand production came from these new wells. Several sand production and erosion studies were initiated to investigate this sand production. Sand transport studies were conducted as part of the investigation. These studies demonstrated that the sand was produced by the 1st well group, rather than the 2nd, but that the sand was not being transported through the main export lines until after the 2nd well group came online and the gas velocities through the flowlines increased. These findings were supported by other observations made in the field following the initial detection of the sand production, while the studies were on going.

Erosion calculations, covering the full range of flow conditions that are expected over the life of the field, were conducted to examine and determine locations of high particle impact rate. The results of these calculations are used to optimize present and future ultrasonic thickness (UT) measurement programs so that critical locations can be targeted with higher localized resolution. Experimental flow studies were also performed to validate the Computational Fluid Dynamics (CFD) erosion calculation model. This CFD model was in turn used to model those parts of the subsea infrastructure that are most prone to erosion under field operating conditions, the Pipeline End Manifold (PLEM). For these flow studies a scaled down version of the PLEM, was built. The CFD results agreed well with the experimental results indicating that the CFD results are accurate in predicting the location of impacts. This was a very important finding because the CFD modeling is used to examine actual pipe geometries within the well and subsea flowline system for various flow and operating conditions during the life of the wells. The results from this work led to more focused and accurate erosion monitoring.

You can access this article if you purchase or spend a download.