Sand Transport Modeling in Heavy Oil Gathering Network in Orinoco Oil Belt, Venezuela

摘要

It is estimated that 70% of the world's oil and gas reserves are contained in reservoirs where sand production is or will become, a problem during the life of the field (Chen et al., 2010). Consequently, an effective sand management strategy may be critical to assure hydrocarbon production. This study presents some results of a developed sand prediction model that has been used in Venezuela to understand sand transport characteristics in heavy oil and to estimate suspension and deposit critical velocities. In addition, results are presented of a sensitivity analysis of particle diameter and water cut carried out using a dynamic multiphase flow simulator to determine the accumulated solids content in some pipelines that helped to develop adequate cleanup procedures. Pressure/volume/temperature (PVT) analyses were initially conducted in order to to reproduce the field characteristics of the produced fluid, including diluted fluid. Sensitivity studies were done to evaluate the effects of some parameters, such as grain size, flow rate, and water cut, to determine how they affect critical transport velocities. From these studies, the volume of particles deposited and the thicknesses of these deposits were determined, which helped the operator to define appropriate pigging program to remove sediments and to estimate the effect on the production system without an appropriate cleanup activity. The initial result shows that the network gathering system operates below deposition critical velocity, however, a stationary sand bed is growing in pipelines near the wells. In addition, the parametric studies revealed that when the particle sizes increase, the critical velocities increase. Besides that, critical velocity shows different behavior with water cut. Critical velocity increases when the water cut goes from 0% to 5%, but if this maximum value is overcome, the critical velocities decrease. Field data indicated that the amount of material received at the end of the system (CPF) during 500 days is 1600 tons of sand, but the maximum operation allowable pressure is reached 290 days after starting up the oil production. Dynamic flow simulations indicated that it is necessary to start the cleaning operation between 150 and 290 days after the start of production, depending on the available pressure to push the pig. This paper summarizes the novel contribution of using dynamic flow simulations for sand prediction and a control model in one of the growing joint venture companies in FAJA PetroIndependencia in Venezuela. Prediction of critical flow rate to prevent sand settling is important for flowlines that are in the design stage. This paper offers a valid approach to extend the predicted sand critical velocities to other fields in FAJA with similar crude conditions to aid in pipeline design.