Sand Production Evaluation: Experimental Testing, Analytical Solution, Numerical Simulation and Field Implications

Abstract

ABSTRACT Sand production prediction is essential from the early stages of field development planning for well completion design and later for production management. Unconsolidated and weakly consolidated sandstones are prone to fail at low flowing bottom hole pressures during hydrocarbon production. To predict the sand-free drawdown, a robust sand production prediction model that integrates near-wellbore and in-situ stresses, rock mechanical properties, well trajectory, reservoir pressure, production and depletion trends is required. Sanding prediction models should be calibrated with field data such as production and well tests observation. In the absence of field data, numerical techniques can provide a reliable estimate on potential onset and severity of sanding at various reservoir pressures. In this study, analytical and finite-element numerical models are independently used to predict the onset and extent of sanding in a high-rate gas well with poorly consolidated sandstone reservoir. The analytical method uses a poro-elastic model and core-calibrated log-derived rock strength profiles with an empirical effective rock strength factor (ESF). In the study, the ESF was calibrated against documented field sanding observation from a well test extended flow periods at the initial reservoir pressure at low drawdown pressure. This sanding model ESF was then verified against the sanding observation from another well test with higher drawdown pressure as well as with the sand-free production conditions from two producers that have been on line for several years. The numerical method uses a poro-elasto-plastic model defined from triaxial core tests. The rock failure criterion in the numerical method is based on a critical strain limit (CSL) corresponding to the failure of the inner wall of thick-walled cylinder core tests that can also satisfy the existing wells sanding observations in the production well tests. To validate the onset and severity of sanding predicted by the analytical model, numerical simulations for an identical sandstone interval were developed to investigate the corresponding CSL. Results showed the numerical model is also able to predict the onset of sanding observed during well tests with plastic strains exceeding the CSL defined from TWC test simulations. This combined analytical and numerical modelling calibrated with field data provided high confidence in the sanding evaluation and their application for future well completion and sand management decisions. The analytical model was finally used for sanding assessment over field life pressure condition because of its processing simplicity, speed and flexibility in assessing various pressure and rock strength scenarios with sensitivity analysis over the whole production interval in compared with the numerical method which is more suitable for single-depth and pressure condition modelling.