The complexities involved in predicting well performance prior to drilling have been one of the greatest challenges faced by the industry since its inception. Every well dwells in its own unique environment and formation and, therefore, presents fresh challenges.

Yet accurate prediction of well inflow is at the crux of the industry’s over-arching goal — to maximize oil and gas production.

Accuracy is vital to appraising development prospects, well planning, and reliable prediction of true well and field value. Prior knowledge about the outcome of proposed actions and well design allows informed choices to be made on formation damage impact and mitigation as well as fit-for-purpose well drainage architecture strategy. But conventional methods lack accuracy and clarity because they have been too cumbersome and imprecise from the outset.

A 90º sector of a vertical well model shows the difference in pressure profile obtained at the fracture plane and at the performance plane. (Images courtesy of Senergy)

Although the quantity and quality of laboratory formation damage testing has increased in recent years and understanding of formation damage mechanisms has improved, translating from pore-scale damage mechanisms to reservoir-scale wells remains extremely difficult. This is because much of the detailed information available from core testing, logs, well testing, and production logging tests is not adequately captured in conventional well inflow models, and their value is greatly diminished by trying to fit the detail into general analytical solutions.

As a consequence, people have to resort to developing boost, fudge, and skin factors to match predicted analytical well performance to real well performance. But by losing the detail of the reservoir, well, completion, and formation damage, the potential for well performance prediction along with predicted productivity indices to be accurate is lost.

Due to the factors that convolute conventional methods, there is a clear need for a flexible, detailed numerical model to predict well inflow.

Hydraulic fracturing is a technique commonly used to mitigate formation damage and improve well productivity. However, the prediction of the production performance of hydraulically fractured wells — an essential step during hydraulic fracture designs — uses analytical solutions that cannot reproduce the complexities associated with flow from the near-wellbore area into the well.

The productivity contrast between the pressure profile for a conventionally drilled (left) and underbalanced horizontal well (right) is a factor of 20.

Despite some recent improvements, the basis of the hydraulic fracture design is contradictory to the real objective, which should be to develop a model to evaluate the fracture performance based on its location and size, then to try to create that optimum fracture using the reservoir and rock mechanical properties. The objective of the hydraulic fracture is to improve productivity (or injectivity), and this must be prioritized.

What is required is a model that can capture the geometry of the fracture or fractures and position them within the reservoir model together with the conductivity of the fractures and the reservoir around the fractures.

Meeting the need

Recognizing the importance of these industry needs, Senergy has developed a proprietary tool, Wellscope, which is targeted to change the industry’s approach to inflow performance and, as a result, increase productivity and cost-effectiveness.

Wellscope addresses the drilling of a well from an entirely different perspective and one that ultra-magnifies the detail of a well’s environment. Based on computational fluid dynamics (CFD), the tool produces a 3-D model of a prospective well to estimate the inflow performance of horizontal, deviated, and vertical wells under various formation damage and completion scenarios.

It has a capability of 10 million cells, which can be recalculated individually or collectively at the touch of a button. This provides not only an insight into how to best evaluate and approach the near-wellbore inflow performance of a well but, as a result, also maximizes the well’s performance.

Michael Byrne, principal formation damage consultant for Senergy and a Society of Petroleum Engineers (SPE) Distinguished Lecturer, said, “CFD is a computational technology that enables study of the dynamics of things that flow. Using CFD, it is possible to build a computational model that represents a system or device. This could be a Formula 1 car where the impact of fluid flow over the surfaces of the car is converted into downforce, or it could be for well flow prediction. The flow of fluid in porous media and pipes and

its restrictions are calculated and compared for different drilling and completion options. The fluid flow physics and chemistry are applied to this virtual prototype, and the software will output a prediction of the fluid dynamics and related physical phenomena.

“Very complex, non-linear mathematical expressions that define the fundamental equations of fluid flow, heat, and materials transport are incorporated,” Byrne said. “These equations are solved iteratively using complex computer algorithms embedded within CFD software. In typical near-wellbore inflow prediction models, 1 to 10 million discrete volumes are used to capture the well and completion geometry and the near-wellbore part of the reservoir. The flow of fluid from the reservoir through each of these volumes into the well bore is calculated simultaneously, and often 200 iterations of the calculation are required to reach a converging solution.”

Maria Jimenez, principal production technologist, evaluated different completion geometries and solutions for a multilayered reservoir. Using CFD modeling of potential productivity, she studied perforation and hydraulic fracture placement. The individual perforations and hydraulic fracture elliptical shape are included in the model, producing considerable detail and reality. Some

of her observations include the productivity improvement in perforations (shot after fracture placement), even in planes where the fracture does not have any direct connection with the well bore. This process has never been captured previously.

“Vertical and deviated well options with multiple fractures were evaluated, and the vertical well option gave the optimum performance. Interference between different fractures, potential breakthrough of pressure profile into unwelcome water or gas legs, and the limitation of deviated well hydraulic fractures were captured,” she said. “Some previous simple well inflow models of hydraulic fractures have relied on analytical solutions to ‘view’ the fracture as a bigger well bore, or they have modeled the fracture as a planar structure. The CFD model captures the true fracture geometry and has thousands of cells on the fracture face (some previous models have just one cell to represent the fracture face).”

This is the first time that CFD has been employed to predict well performance based on high-quality laboratory testing and the first time it has been used as a practical tool to assess the well drainage architecture design of hydraulic fractures. It is anticipated that the modeling of well inflow in great detail and in three dimensions using CFD will become routine in the next few years.