Mitigation of sand production is increasingly becoming an important and challenging issue in the petroleum industry. This is because the increasing demand for oil and gas resources is forcing the industry to expand its production operations in more challenging unconsolidated reservoir rocks and depleted sandstones with more complex well completion architecture. A sand production prediction study is now often an integral part of an overall field development planning study to see if and when sand production will be an issue over the life of the field. The appropriate type of sand control measures and a cost-effective sand management strategy are adopted for the field depending on timing and the severity of predicted sand production.
This paper presents a geomechanical modelling approach that integrates production or flow tests history with information from drilling data, well logs and rock mechanics tests. The approach has been applied to three fields in the Australasia region, all with different geological settings. The studies resulted in recommendations for three different well completion and sand control approaches. This highlights that there is no unique solution for sand production problems, and that a robust geomechanical model is capable of finding a field-specific solution considering in-situ stresses, rock strength, well trajectory, reservoir depletion, drawdown and perforation strategy. The approach results in cost-effective decision making for appropriate well/perforation trajectory, completion type (e.g. cased hole, openhole or liner completion), drawdown control or delayed sand control installation. This type of timely decision making often turns what may be perceived as an economically marginal field development scenario into a profitable project.
This paper presents three case studies to provide well engineers with guidelines to understanding the principles and overall workflow involved in sand production prediction and minimisation of sand production risk by optimising completion type.
Review of Sand Production Prediction Models