Conventional well inflow modeling fails to capture the complexity of flow in to wells from asymmetric near wellbore permeability distribution. In order to simulate flow in to horizontal wells with varying degrees of damage along the well and around the circumference a complex numerical model is required. Computational fluid dynamics enables fluid flow in to the well to be modeled. Along a horizontal well the magnitude and depth of mud cake, filtrate invasion and formation damage will vary. Longer exposure times at the heel of a well and lower drawdown at the toe will lead to different formation damage, clean-up and inflow patterns along the well. In addition, gravity can have a profound effect on damage distribution at the base, middle and top of the well. Thicker mud cakes will tend to develop at the base of the well and whilst there may be more erosion on the low side, this will only serve to compound the damage by leading to deeper mud invasion. Inevitably, formation damage will have an asymmetric distribution along and around the well. Asymmetric models of fluid flow in to horizontal wells have been created to reflect the damage heterogeneity. Results of the models have shown that even in homogenous reservoirs, the vast majority of inflow in to the well is derived from the high side of the heel of the well. Modeling shows the potential impact of consistent chemical clean-up and of inflow control along the horizontal well. By fully capturing the magnitude and impact of damage in conventional wells using computational fluid dynamics, the value of well drilling and completion options such as underbalance drilling and hydraulic fracturing can be evaluated. This paper presents the first use of complex three dimensional computational fluid dynamic models to predict and mitigate against asymmetric formation damage in horizontal wells.

Computational Fluid Dynamics (CFD) has been used to model fluid behaviour in reservoirs, near wellbores and wells. More recently the CFD has been applied to more diverse challenges such as sand transport, coupled mechanical modelling of failing wellbores and gas well clean-up. CFD has been applied to model the crushed zone around perforation tunnels and also to predict well performance base on laboratory testing. Where the reservoir is involved in these models they are fully coupled and flow is enabled deep in the reservoir and in to every point along the well. Capturing the simultaneous flow of fluids in the reservoir and in the well is essential to predict performance in wells and reservoirs with complex geometry. Cross flow between reservoir layers and the flow of fluids through and along the well length are real phenomena which if ignored, can lead to poor prediction of well performance. CFD modelling relies on fundamental physical properties and enables changes in flow restriction such as formation damage or matrix stimulation to be captured. This paper presents some recent case histories where model results are verified by real well performance and where traditional analytical model results are compared to fully coupled CFD reservoir and well models. The impact of formation damage is observed to produce very different results, depending on the model applied. Case histories illustrating the application of innovative, rigorous modelling of formation damage from different drilling and completion fluids and practices are presented. Further advances that enable sand failure and sand transport to be modelled are also illustrated. The case is made for use of CFD or similar modelling processes where complex reservoirs or complex wells are considered. IntroductionCFD has well tried and tested applications in many industries including the oil and gas industry. Here it has principally been used to model fluid flow in surface pipelines and to examine fluid flow through and around tubing and equipment. For example CFD is used in the design and development of new drill bits and in the development of Inflow Control Devices (ICDs). More recently there has been an increasing trend to move CFD in to the well and the complexity of completion and reservoir geometries (Byrne et al 2010). This evolution has been enabled by the advance in computational power and speed and by improvements in CFD software packages. CFD enables modeling of the physics of fluid flow through restrictive media and so is ideally suited to modeling the flow of fluid or fluids through reservoirs, completions and wells. Previous papers have presented some of the earliest work in this field but even in a few short years significant progress in the accuracy and the scale of modeling has been made. Complex phenomena such as unloading liquid completion fluid from gas wells, sand failure and the impact of changing wellbore shape on fluid flow and complex well geometries have now been modeled and used to assist in the design of optimum wells. A case study is pr...

During coring operations, while tripping out of the hole, core is submitted to a relatively sudden decrease in pressure, leading to fluid expansion and movement out of the pore space. The rapidly expanding fluids can generate fractures and damage the core. Tripping out core represents a delay to well drilling operations, therefore, project economics dictate "efficient" tripping out schedules given high daily rig costs. Ideally, core tripping out rates should be established for each particular core; however common practice is based on generic rules of thumb. The resulting schedules can lead to unjustifiably long tripping out operations or core damage if the process is too fast. There is a lack of clarity and consensus regarding tripping schedules impacting on both the integrity of the core and the economics of coring/drilling operations.With these high daily rig costs in mind, a more scientific and quantitative approach, tailored to each case is required. The application of Computational Fluid Dynamics (CFD) is used to model transient pressure differentials in a gas reservoir core during retrieval.The results of theoretical and field case simulations support and confirm empirical evidence that the pressure differentials created in a core during retrieval, are very low for relatively high permeability rock, and that most core tripping schedules based on rules of thumb significantly overestimate time, to the detriment of drilling efficiency. As the core permeability decreases, the time required for the core internal pressure to approach the external pressure increases, requiring longer equilibration times. The model realisations demonstrate that tensile failure criteria are more likely to be reached during the final stages of tripping out.This study demonstrates that CFD can accurately predict the pressure differentials created in a core during retrieval to surface and enables proper planning of tripping times based on the assessment of potential damage by pressure release.

Formation damage sometimes has a significant impact on well productivity or injectivity. Much of the work undertaken to mitigate against formation damage serves to qualify the impact of damage and for example, determine the least damaging introduced fluids. Whilst qualification is important, quantification of damage impact is a much greater prize. Taking advantage of increased computational power and speed, Computational Fluid Dynamics (CFD) has been applied in the last five years to numerically model and quantify flow restriction in wells and the near wellbore (Byrne et al 2009). This process enables better prediction of the real impact of formation damage on well productivity. The process has been applied to many real fields and wells and in this paper we present some of the latest revelations on the impact of formation damage on well performance. In order to illustrate the simplicity of the process a sensitivity analysis of typical formation damage restrictions has been applied. This demonstrates the impact of different formation damage quantities and depths on different well trajectories and lengths in the same reservoir. In some cases the impact of damage that could be classed as "severe" in a laboratory test has little impact on eventual well productivity. In others the damage has a profound impact on well performance. With the obvious caveat that every well is different, we propose some consistent guidelines on the impact of damage based on our numerical modelling work. Modelling of this kind does not reduce the necessity to understand or simulate damage, instead it enables focus where formation damage has greatest impact and creates a greater awareness of the importance of damage to our industry. Through increasing use of appropriate modelling, uncertainty is reduced and better wells can be drilled and completed.

Gyda is a mature oil development in the Norwegian sector of the North Sea. The first production wells were drilled more than twenty years ago. This study focuses on wells drilled in the porer reservoir quality areas of the Gyda reservoir. Some recent production wells have significantly underperformed relative to equivalent initial wells. In particular, a sidetrack to an early successful well, had very poor performance on initial start-up.The geometry of the original well and the sidetrack were simulated, together with various assumptions and sensitivities to formation damage. In the original well an attempted hydraulic fracture had been assumed to have failed. This assumption was challenged in the model.The model demonstrated that the original well must have included a successful hydraulic fracture in order to flow at the historical rates recorded. In addition for the sidetrack, that contained no fracture, there were indications that the perforation tunnels may not have fully cleaned up and that whilst the well performance may recover somewhat with time, a significant change in completion would be required in order to match the performance of the original well.The model constructed included the completion geometry and formation damage and has enabled evaluation of old wells and more importantly, design of new wells in this mature reservoir development. IntroductionSome recent wells drilled on Gyda have not fulfilled the production objectives. A numerical 3D model was proposed in order to investigate and understand the flow dynamics and the production potential from the Gyda A19 and A19A wells. This modeling process includes a detailed numerical fluid flow simulator based on Computational Fluid Dynamics (CFD) which captures the reservoir, well and completion geometry complexity (Byrne et al, 2009 and2010).The CFD simulations are used to determine potential explanations for the wells performance and lead to stimulation options and development of optimum drilling and completion for future wells. The Senergy Wellscope modeling process has the stated objective of better production through better prediction. As the A-19 well is very similar to the A-19A well from Gyda, some conclusions may be derived from the present study which could support the understanding of the productivity behaviour of the A19A well.To achieve the objective one base CFD model was constructed to represent the two wells to be evaluated (A19 and A19A). Different completion options were included, including the case of the well hydraulically fractured. Several sensitivities were carried out in order to depict the well potential.Throughout the project, regular contact was held between the Senergy and Gyda team. These helped to frame and direct the project as well as providing necessary feedback on data gathered and model construction.