Complex Completion Design and Inflow Prediction Enabled by Detailed Numerical Well Modeling

Abstract

An Operator was unable to model a potential new (additional) completion of a 330 m cased hole perforated sand screen and ICDs interval in a mature high water cut well which was originally completed with a 916 m open hole stand alone sand screen and Inflow Control Devices (ICDs) section in a multidarcy sand. The potential opportunity to perforate this 330 m section presented significant potential reservoir drainage upside but could not be modelled using conventional well inflow prediction or reservoir simulation techniques. In order to determine if the recompletion was economically viable, the operator required a way to model the recompletion and the existing completion to determine the overall completion performance. The complexity of the original open hole section completed with sand screens and ICDs and the new target to be completed with perforations, sand screens and ICDs was solved using computational fluid dynamics (CFD) modelling and high performance computing. The existing and new reservoir intervals are characterised by unconsolidated high permeability sands. The reservoir conditions mobility ratio of oil to water is approximately 30:1. Due to high total liquid production rates, the existing open hole completion is producing at well above the economic oil rate cut off despite being at approximately 95% water cut and therefore the existing completion interval cannot be abandoned. The new recompletion perforation interval would initially produce at 100% oil. The key question was, what will the new recompletion interval add (if anything) to the overall well production rates and is the new recompletion economically viable. Conventional analytical or even 1D or 2D numerical models simply cannot handle the complexity of the geometry of this well's open and potential cased hole intervals, perforated intervals, sand screens and ICDs. A 3D fully coupled well model was constructed and 2 phase CFD modelling undertaken in a combined model size of over 500 million cells each with unique properties. Through employment of what is thought to be the most comprehensive inflow model ever built, the contribution from the original open hole interval and the new interval were estimated and the optimum completion design investigated allowing the operator to determine the economic viability of the recompletion.

A new assessment of perforation skin factor for vertical perforated wells in near-wellbore region

“…Also, Byrne et al (2010 and2011) demonstrated how CFD could resolve asymmetric distribution formation damage near the well and capture the cross flow in the layered reservoir with a heterogeneous. A few years later, the same researchers developed a model for a perforated horizontal well (Byrne et al 2014). Additionally, Molina and Tyagi (2015) applied CFD on a near-well model to examine how a perforated gas well performs, looking in particular at various types of completion techniques and how these strategies might have erosive effects at the wellbore.…”

Near-wellbore modeling of a horizontal well with Computational Fluid Dynamics

“…Byrne showed that CFD can resolve the formation damage of asymmetric distribution around the well (Byrne, et al, 2010), and can capture the cross-flow appearing among adjacent layers for a heterogeneous layered reservoir (Byrne, et al, 2011). Furthermore, a case study was published about the detailed modeling of a perforated horizontal well, offshore Australia (Byrne, et al, 2014). Moreover, Molina (2015) have used CFD for investigating the performance of a perforated gas well using a detailed near-wellbore model, focusing on different completion methods and erosive effects on the wellbore (Molina, 2015).…”

Development and investigation of integrated reservoir-wellbore flow model

“…Other methods of investigating the performance of perforated tubes involve theoretical and numerical simulation. Some studies have used CFD modelling to overcome the abovementioned difficulties for a wider range of applications (Byrne et al 2009;Sun et al 2011;Byrne et al 2014;Molina and Tyagi, 2015). Su (1996) used the commercial CFD code, FLUENT to simulate the smoothing effect of the wall inflow.…”