Permian Basin: Use of In-Situ Mechanical Rock Properties Improves Completions
You have access to this full article to experience the outstanding content available to SPE members and JPT subscribers.
Optimizing horizontal well placement is not limited to identifying the most-favorable reservoir, but also involves identifying the ideal target window within that reservoir. Gathering drill-bit geomechanics data provides a lower-cost and lower-risk method to acquire mechanical rock properties in long horizontal wellbores. By integrating data sets with mechanical rock properties recorded while drilling, operators can have significantly higher confidence in choosing a target landing zone and improving completions. The complete paper presents two detailed case studies from the Permian Basin.
In this paper, the authors combine the characterization of petrophysical and geomechanical properties into what they call a petromechanical work flow. Typically, petrophysical and geomechanical properties are characterized using data acquired by wireline logs. In vertical wells, wireline logs represent, traditionally, the least-intrusive manner of acquiring high-resolution data. With additional cost and rig time, cores of the formation rock can be exhumed to analyze its properties on surface. In horizontal wells, both wireline logs and conventional cores are costly and operationally challenging. Because of the economic and operational burden, operators typically avoid collecting geomechanical data in horizontal wells. Besides these traditional measurements, the novel technique of using drill-bit geomechanics can enable measurement of geomechanical properties while drilling. Compared with wireline logs and core analysis, this technique is less costly and has a lower risk in long horizontal wells.
Drill-bit geomechanics provides mechanical properties through continuous, high-resolution measurements of drilling-induced vibrations. Triaxial accelerometers, which sample at 1 kHz, record the vibrations while positioned directly behind the drill bit. Earthquake seismology models allow a transformation of the high-frequency, triaxial, drilling-induced vibrations into mechanical rock properties.
The drill-bit geomechanics method determines mechanical properties by using stress-strain relationships and isotropic stiffness coefficients. Additionally, the method describes anisotropy by solving for transversely isotropic (TI) interpretations of the rock matrix. Typically, a vertically transverse isotropic (VTI) matrix contains high shale content or other laminar bedding, while a horizontally transverse isotropic (HTI) matrix contains vertical bedding or fractures. In this paper, the authors define VTI anisotropy as bedding and HTI anisotropy as fracture intensity.
Case Study: Pilot Well
Comparison of Multiple Measurements. For mechanical properties, an operator can use drill-bit geomechanics data to supplement sonic log data. Sensitivity of the latter to fluids and pressure can be leveraged to achieve a detailed reservoir analysis. The nonuniqueness of sonic measurements with respect to fluid content, pore pressure, and the mineralogy of the matrix results in uncertainty that can affect horizontal development decisions, as well as geological reservoir characterization. Some discrepancy between the high-resolution mechanical curves and the sonic curves is expected because the sonic logging tool has approximately a 4-ft resolution and the drill-bit geomechanics method has less than 6 in. of resolution. Also, dynamic factors have a greater effect on sonic measurements than on drill-bit geomechanics.
Drill-bit geomechanics data are typically much higher resolution than are those derived from sonic logs. To evaluate the validity of these high-resolution changes, the study team compared them directly with core photographs. The mechanical curves successfully identified lamina-scale changes on the core photo.
Like drill-bit geomechanics and core, wireline image logs can measure with exceptionally high resolution.
Petromechanical Work Flow for Target-Zone Identification. A well-established petrophysical work flow involves grouping similar rock types into lithofacies. The same facies modeling can be applied to the drill-bit geomechanics data to create mechanical facies. The combination of multiple measurements of varying degrees of resolution can improve an operator’s decisions about horizontal well development.
Case Study: Three Horizontal Wells
An operator drilled three neighboring horizontal wells in the same target landing zone: Wells A, B, and C. The wells, spaced approximately 660 ft from one another, had approximately 10,000-ft parallel laterals. While drilling, the operator acquired drill-bit geomechanics data along the well paths and used the data to design each well’s completion. Despite some minor differences in perforation cluster spacing and stage length, the three wells shared a similar completion design overall. All three designs used the same number of perforation clusters per stage, treatment fluid, proppant size, proppant concentration, and treating rate.
Using step-down rate tests, the engineers validated that all three wells’ perforation and stage designs were equally effective. The average perforation efficiencies (i.e., percent of open perforations) for Wells A, B, and C ranged from 62 to 67%. Similarly, the average calculated tortuosity values differed by less than 15%. Because step-down rate tests occur at the start of each stage, before proppant is pumped, these results suggest all three wells’ completions were equally well-designed.
Despite their geographic proximity and their similar completions, the wells showed notably different production responses.
Petromechanical Work Flow for Completions. For completions, the petromechanical work flow leverages petrophysical resources, applies them to mechanical data, and provides a predictive framework for mechanical behavior. The mechanical behavior investigated in this case study is that of hydraulic fracturing. The ability to hydraulically fracture effectively depends upon mechanical quality. To approximate mechanical quality, the work flow uses mineralogic brittleness and unconfined compressive strength (UCS).
Designed for use in horizontal wells without extensive wireline log data, the petromechanical work flow estimates basic mineralogy from the measurement-while-drilling gamma ray and drill-bit-geomechanics-derived bedding (i.e., VTI anisotropy). Clay volume can be calculated from gamma ray and bedding independently.
From the basic mineralogic model, the work flow calculates brittleness. Mineralogic brittleness is a linear function of the bedding-derived clay volume. The work flow calculates UCS from the drill-bit-geomechanics-derived Young’s modulus.
Fig. 1 shows the mineralogic model and the calculated UCS and brittleness for each well. The petromechanical work flow displays UCS and brittleness on the same track (Track 2) to illustrate their relationship to one another. The authors consider intervals of relatively high brittleness and low strength as high mechanical quality. Thus, these intervals have a positive crossover, shown in green. On the other hand, values of relatively low brittleness and high strength have a negative crossover, shown in grey. The authors consider intervals of high mechanical quality as generally easier to fracture and prop, which would result in a more-conductive fracture network.
As shown in Fig. 1, Well A has a greater percentage of high mechanical quality than Wells B and C. Approximately 75% of Well A’s lateral has high mechanical quality compared with approximately 25% of Well B’s lateral and approximately 25% of Well C’s lateral. These observations are consistent with the production responses. Of the wells with consistent production mechanisms, Well A is the best producer in the development area. After 100 days of normalized production, Well A produced approximately 30% more oil than Wells B and C. The authors hypothesize that Well A’s hydraulic fracturing treatment created a better fracture network than those of Wells B and C because of A’s superior rock quality.
Because the operator performed step-down rate tests before each stage, initial shut-in pressures (ISIPs) were collected before pumping proppant. All three wells were completed in a zipper fashion. Thus, the prestage ISIP measurement is not a perfect comparison to near-wellbore mechanical properties because it can contain effects of interstage and even interwell stress shadowing. Despite these caveats, the engineers found a general trend between mechanical quality and prestage ISIP. Fig. 2 shows the mechanical quality curves for each well with the prestage ISIPs. The ISIPs were converted to gradient to normalize by depth. Prestage ISIPs are represented by bar graphs in which higher bars correspond with higher ISIPs. In general, intervals with higher mechanical quality (i.e., green shading) have lower ISIPs, while intervals with lower mechanical quality (i.e., grey shading) have higher ISIPs. Assuming that high prestage ISIPs indicate reservoir rock that is more difficult to fracture, this correlation validates the authors’ hypothesis.
The first few stages of the well are expected to have lower ISIPs because of limited stress-shadowing effects. In these toe stages, the ISIPs increase gradually as each stage introduces more pressure and greater stress to the formation. The variation in ISIPs for toe stages is likely driven by completion order, not mechanical properties. After these first few stages, ISIP variance is likely driven by mechanical quality. Note that in all three wells, the heel stages have the greatest percentage of high mechanical quality and that, in all three wells, ISIPs are lowest in the heel, shown on the left side of the image. From the toe, on the right side of the image, Well A has lower mechanical quality in the first approximately 25% of the well and Well B has consistently lower mechanical quality through the first approximately 40% of the well. In both cases, ISIPs are higher in these intervals, suggesting greater difficulty for fracturing. Well C has sporadic intervals of high and low mechanical quality and its ISIPs also fluctuate sporadically.
In a pilot well, drill-bit geomechanics data compare well with, and complement, the mechanical properties derived from more-traditional reservoir-characterization tools such as wireline logs and core. Combining these measurements can inform the best decisions for landing zones and horizontal well development. The case study of the three neighboring horizontal wells with notably different production highlights the importance of geomechanical understanding. Blindly extrapolating geomechanical properties from a vertical pilot well to all nearby horizontal wells can cause inaccurate assumptions and interpretations. Drill-bit geomechanics offers an economical and operationally undemanding way to measure mechanical properties along horizontal wells.
Permian Basin: Use of In-Situ Mechanical Rock Properties Improves Completions
01 November 2019
Horizontal and Extended-Reach Wells
Drilling has become a complex operation that requires a multitude of tasks to be managed at the same time. Not surprisingly, then, drilling automation will gain momentum in the near future to deliver consistent execution and performance safely and with good wellbore quality.
Nonaqueous, Nondamaging Fluid Implemented Offshore Abu Dhabi
This paper covers the 7-year history of drilling-fluids application in a reservoir drilling campaign offshore Abu Dhabi, from the early use of a solids-free, brine-/water-based mud to the recent application of nondamaging, nonaqueous fluids (NAFs) with micronized acid-soluble ilmenite.
Reduced-Friction Centralizers Field-Tested for Ultraextended-Reach-Drilling Completions
The complete paper documents the development, laboratory- and field testing, and lessons learned from a project to evaluate coated centralizers.
No editorial available
Don't miss out on the latest technology delivered to your email weekly. Sign up for the JPT newsletter. If you are not logged in, you will receive a confirmation email that you will need to click on to confirm you want to receive the newsletter.
06 November 2019
05 November 2019
07 November 2019
05 November 2019
14 November 2019
No editorial available