Optimizing horizontal wells is one of the major contributing factors to a successful economic recovery of unconventional reservoirs. That is why operators continue to seek new solutions for improving various completion parameters that directly impact well productivity. By conducting completion design pilot tests, operators can determine the most appropriate number of fracturing stages, cluster spacing, fracture design and other critical parameters affecting completion performance. However, such tests can be both cost- and time-prohibitive and might not provide much-needed answers.

One way to reduce costs and speed up completions is through numerical modeling of completion designs via cloud-based computing. The most valuable benefit of this method is that decision makers can study and understand a large number of variable samples rapidly and direct their field operations based on the assessment of numerous what-if scenarios—all of it accomplished in real time. This has a direct impact on production enhancement as numerical modeling simulations enable more accurate mapping of reservoir heterogeneity, more precise characterization of reservoir quality and a more defined process of selecting and placing effective completions in the wellbore.

As seen in a recent Wolfcamp Shale case study, hundreds of modeling simulations are required to understand trends in hydraulic fracture geometry and productivity when developing the most suitable completion design plan for an unconventional asset. This is only practical when an automated workflow is powered through cloud-based parallel simulations that thread the hydraulic fracture design, unstructured gridding and numerical simulation for production response.

Integrated earth modeling

In the Midland Basin of the Wolfcamp Shale, cloud computing techniques played a crucial role in optimizing well completion and spacing design of a multiwell pad. As a first step, creation and calibration of a 3-D earth model on the Petrel E&P software platform took place representing the asset’s geological, geomechanical and petrophysical properties. After these properties and the reservoir’s discrete natural fracture network were defined, cloudbased computing was used to perform a multivariate analysis to optimize the well completion design and well spacing. The following completion parameters were used:

  • Proppant loading: 1,000 lbm/ft to 5,000 lbm/ft;
  • Cluster spacing: 6 m to 38 m (20 ft to 125 ft);
  • Number of clusters per stage: 3 to 7; and
  • Horizontal well spacing: 91 m to 305 m (300 ft to 1,000 ft).

Additionally, petrotechnical experts used the Kinetix Shale reservoir-centric stimulation-to-production software to understand fracture geometries for zipper and nonzipper stimulation sequences and the effects of existing well production on reservoir geomechanical properties and infill well productivity. Several critical indicators of production and hydraulic fracture geometry parameters were evaluated, such as total and propped surface area; height, length and width of the fractures; and net pressure in the fracture.

Simulation engines

By using the numerical modeling approach, more than 500 cloud-based complex simulations of hydraulic fractures, as well as unstructured gridding of hydraulic fractures with fine-resolution numerical and finite-element geomechanical simulations, were performed to determine

  1. An optimal well landing solution by using a full 3-D hydraulic fracture simulation model and complex fracture models in the Kinetix Shale software;
  2. Simulated values to match with field measurements, such as treatment pressure history, microseismic data and production history. They provided calibration points for hydraulic fracture geometry and productive reservoir volume representation;
  3. Future well performance for all completion sensitivity cases. Cloud-based simulations using the INTERSECT high-resolution reservoir simulator were implemented to predict this performance; and
  4. Parent-child well relationship and the effect of stimulation timing on child wells. These parameters were established by using the VISAGE finite-element geomechanics simulator to predict reservoir geomechanical property changes over time.

Achieving these results through conventional computing workflows—such as manual, single simulation at a time—would have taken months to years. Instead, the numerical, cloud-performed methodology delivered the results within a week.

Proppant loading and perforation clusters

Production increases with stimulation treatment size—but up to a certain level. Cloud-based simulations of the 3-D earth model have shown the total generated fracture surface area improves when increasing volume of proppants, with the propped surface area plateauing at about 3,000 lbm/ft. This has enabled a faster and more accurate economic analysis of the resulting production to determine the optimal proppant loading.

Optimizing cluster spacing when completing a well is another technical challenge in this region. The cloud computing simulations using Kinetix Shale software were analyzed showing that as the cluster spacing is reduced, more near-wellbore complexity and interaction with the natural fractures result in increased productive surface area.

The analysis also demonstrated that as the clusters per stage increase, the fracture length drops because the fluid volume per cluster falls. However, the resulting surface area, fracture height, fracture conductivity and fracture width do not change significantly. Hence, the number of clusters per stage has less effect on well productivity as compared with proppant loading and cluster spacing. In other words, modeling demonstrated that the operator can improve production and overall project economics by reducing cluster spacing instead of increasing proppant loading (Figure 1).

FIGURE 1. The Wolfcamp study indicated that smaller proppant loading at tighter cluster spacing results in slightly higher production compared with wider spacing and larger proppant loading. (Source: Schlumberger)

Impact of zipper fracturing

Operators use the zipper fracturing technique to improve operational efficiency while stimulating multiple wellbores. In the Wolfcamp case study, the fracture geometry impact of a zipper fracture case was compared to a nonzipper sequential stimulation case on a four-well pad, finding that the interwell stress shadow effect is minimal until the volume reaches 2,400 lbm/ft.

Well spacing

In a multiwell pad, tighter well spacing usually results in fractures competing for the same rock volume; therefore, production interference is commonly observed. Here, however, marginal to no production interference occurs at 200-m (660-ft) well spacing over a twoyear cumulative production. Nevertheless, production interference increases to approximately 8% at 135-m (440-ft) well spacing and 18% with 100-m (330-ft) well spacing. Also, another finding from this case study is that treatment design can affect the well spacing decision—the larger the treatment, the farther the well spacing should be to mitigate production interference.

Parent-child wells

It is a known fact that existing well production induces a timedependent geomechanical property change that shapes the nearby infill wells’ fracture propagation, fracturing hits and well productivity. Close well spacing between existing and infill or parent-child wells tends to result in a greater number of fracturing hits. This spacing sensitivity generated through cloud spacing analysis of a parent-child system for the Wolfcamp asset indicated that at a spacing of 135 m and closer, the probability of a fracture hit is significantly higher than for a system at 200-m (660-ft) spacing (Figure 2).

FIGURE 2. A multivariate analysis of a Wolfcamp asset determined that increasing proppant loading optimizes production in an infill well but also likely results in fracturing hits with a detrimental effect on production from a nearby existing well. The cloud-based study determined that reducing cluster spacing in the infill well optimizes overall project economics. (Source: Schlumberger)

Bottom line

By applying the cloud-based reservoir modeling and simulations, the operator was able to place more wells per section, increased productivity per well by more than 40% and improved the net asset value by more than 50%. As exemplified in this case study, time and cost savings can be achieved through a cloud-based sensitivity study for operators who strive for optimized completion design. Booking reserves, economic evaluations and field trials can be completed with optimal assurance and in a short time frame.

Editor’s Note: This article has been adapted from the URTEC- 2876482 and SPE-191442-MS papers, both 2018.

Have a story idea for Shale Solutions? This feature highlights technologies and techniques that are helping shale players overcome their operating challenges. Submit your story ideas to Group Managing Editor Jo Ann Davy at jdavy@hartenergy.com.

Comments