SPE-187830-MS

Application of 3D and Near-Wellbore Geomechanical Models for WellTrajectories Optimization

D. V. Alchibaev, A. Ye. Glazyrina, Yu. V. Ovcharenko, O. Yu. Kalinin, S. V. Lukin, A. N. Martemyanov, S. V.Zhigulskiy, I. S. Chebyshev, and A. V. Sidelnik, OOO Gazpromneft NTC; I. Sh. Bazyrov, Saint-Petersburg MiningUniversity

Copyright 2017, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Russian Petroleum Technology Conference held in Moscow, Russia, 16-18 October 2017.

This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contentsof the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflectany position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the writtenconsent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations maynot be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

AbstractFor the prediction and elimination of complications in the drilling process is considered a number ofexamples of the three-dimensional geomechanical model and of the near-wellbore model in order tooptimize the trajectory and design of the wells. During the well trajectory planning, the key point isto forecast and minimize all possible risks associated with both geological, mechanical conditions andtechnological parameters. An optimal solution can be obtained with the use of a detailed geomechanicalanalysis.

It is shown that in a number of cases, the numerical model of the near wellbore zone is moreinformative, in comparison with the analytical solution. The result of drilling risks minimization with helpof geomechanical analysis tools is presented. A number of recommendations on wellbore construction andstability are established of the comprehensive geomechanical analysis. The discontunities that are derivedof seismic field analysis are also included in the review.

The image analysis, 1D geomechanical modelling, of seismic field analysis, near-wellbore numericalsimulation and full 3D goemechanical modelling were used as a geomechanics tools to optimize "fishbone"trajectory. Microimages help to determine the presence of cavernousness, natural and induced fractures,geological boundaries and bedding planes. Especially useful is a tool for determining the presence ofcollapse in the areas of kick-off sidetracks. 1D geomechanical modeling helps to determine favorableintervals for shearing and optimal mud density. To assess the risks during the sidetracking operation, astatistical analysis of the actual data was carried out taking into account the spatial orientation of thesidetrack and the direction relative to the currently acting stress state. Stresses and gradients of caving inthe intervals of cuts are refined by the near wellbore model.

IntroductionToday oil and gas specialists have to solve the problems of production optimization in order to reduce costsduring the development of fields and increase the cumulative oil production. In terms of drilling the taskof optimization can be expressed as the selection of the optimal trajectory of wellbore during drilling: the

2 SPE-187830-MS

correct choice of the well trajectory parameters significantly reduces costs by preventing accidents andtheir subsequent elimination during drilling and round-trip operations. In addition, for example, a horizontalsection of the well can be selected in order achieve the most favorable conditions for subsequent multiplehydrofracturing of the formation. Moreover, orientation and location of horizontal well can be chosenfollowing the stress-field favorable to conduct multi-staged hydraulic fracturing.

To solve the first of the above-described questions we have to use the method which is based on thewellbore stability estimation which directly depends on the stress state, rock strength, mud weight and thecorresponding equivalent circulating density during drilling, as well as well trajectory. Thus, in terms ofaccidents trajectory optimization is based on the hole drilling in mechanically most favorable sections andchoosing the most stable variant of wellbore geometry – well inclination and well azimuth.

Figure 1—a) structure variations in the interval of Bazhenov formation and overlying and underlyingformations with thinning away; b) map of Bazhenov formation's top; c) faults, indicated in Bazhenov formation

Well-completion technique can edit additional constraints. Today multi-stage hydro-fracture technologyis becoming more and more popular. Geometry of fracture, processing limits, borehole surveys and desirableadditional surveys after hydrofracturing are the factors, which have to be taken into account during welltrajectory optimization. Initiation of injection in improper interval can cause unlimited fracture growth inheight, underestimation of changes in orientation of stresses in the lateral direction can cause reversal of thefracture orientation from the planned azimuth and will result in a decrease in production, increased risks ofwater breakthrough and other undesirable consequences [1].

The key and the first question on the way to solving the above problems is to determine the current stateof stress. Heterogeneities of the structure, such as such as upliftings and deflections, folds, thinnings, dikes

SPE-187830-MS 3

and salt domes, lead to a change in the magnitude and orientation of the main stresses due to significantchanges in the mechanical properties of rocks of different lithological composition [2].

Complex fault tectonics introduces additional perturbations in the stress field and leads to nontrivialmovements both in lateral (block formation, pullapart) and in vertical directions (normal faults, strike slipfault, reverse fault).

The stress-strain state of the rock can also vary due to pore pressure, which can take anomalously high orlow values due to natural or technogenic causes. Pore pressure changes in one formation can significantlyaffect the values and orientation of stresses in adjacent formations and cause the activation of faults andother effects.

Figure 2—Stress field changes under the influence of faults

Near-wellbore stress stateThe most common approach in calculation of near-wellbore stress state is by using an analytical solution forinfinite 2D plate with circular hole under far-field stresses done by Kirsch[3]. The problem has an analyticalsolution within the assumptions of plane strains. This approach is limited by the ideal circular geometry ofthe wellbore with only cross-section that is perpendicular to well direction. In case of vertical well, verticalstress stays the same and horizontal stresses can be calculated by following system of equations:

(1)

4 SPE-187830-MS

In case of deviated or horizontal well, to use analytical solution one must rotate in-situ stress tensor tocoordinate system with one of the axies co-directional to well trajectory.

(2)

Figure 3—Coordinate system rotation

Thus, knowing the stress state near borehole along the well and applying failure criteria – Morh-Coulombshear failure criteria is commonly used – one can predict drilling risks for any given mud weight.

Figure 4 shows wellbore stability break-out model for various well orientations in normal faulting stressstate. Point corresponds to given well trajectory characteristics at given depth (inclination 50 deg, azimuth105 deg) and mud weight 1.5 g/cm3.

Figure 4—Break-out gradient vs. well trajectory for one of geomechanicaly unstablezones. a) break-out gradient vs. well azimuth; b) break-out gradient vs. well inclination

Undeniable advantages of analytical solution approach are simplicity and calculation speed. However,analytical solution contains number of limitations. Analytical calculations are accurate until:

• Borehole walls are intact borehole has a cylindrical shape with circular cross-section;

• Rock near wellbore retains lateral homogeneity;

SPE-187830-MS 5

• In case of deviated or horizontal well – a rather slow change of mechanical properties with depth.

It is very helpful to use near-wellbore modeling for examine stress and strain fields when the soil loadsare more than the Hook area. In this case, inelastic behavior of the soil can be observed. Deformations inthe inelastic area are consisting of elastic and plastic terms. Note, the plasticity effect leads to volume andstrain energy change, also it means that pore volume and permeability change in plastic area.

To show forming drilling induced breakouts a step by step numerical modeling of near-wellbore area isconducted. Let's initially consider the plane strain problem for the circular hole in infinity plane and thefailure of material will occur when the Drucker-Prager function is less than zero. [4]

where α = α(ep, σ) and c = c(ep, σ) are the factors of internal friction and cohesion, respectively;

– is the second stress deviator invariant. A normal to the function contour g = τ - Λσ. indicates thedirection of the plastic strain increment, where Λ is a dilatancy factor. The shear strain increment intensity

is calculated from , and the volume strain increment dεp = -(6εkk)p is

determined from .

Figure 5—Druker-Prager surface projection principal stress planes[4]

As a result of sequential breakout evolution modelling is shown if the wellbore failure starts. Withincreasing the breakouts near wellbore stresses magnitude became less than critical stress, so the boreholestability occurs with deformed form.

Figure 6—Circular stresses in the direction of minimal horizontal stress for circular cross-section for intact and failed wellbore

6 SPE-187830-MS

It is shown circular and radial stresses plots of every modelling phase (Figure 7). In case the circular holeand defined material properties and loads, a failure occurs in 15% of wellbore radius (grey plot). The nextphase (green plot) shows the stress field in near wellbore area has no critical values and wellbore stabilitystate occurs with measured fail.

Figure 7—Circular stresses in the direction of minimal horizontal stress. S_lim – shear failurestress for intact wellbore, S_lim 15% - shear failure stress for failed wellbore (break-outs 15%).

Deviated well trajectory optimizationSeveral parameters are to consider when designing optimal well trajectory: should satisfy geological target,to take into account drill bit parameters, well construction limitations, mud weight, maximum doglegseverity, weight on bit and torque etc. Besides, the reduction in construction costs implies the selection ofthe trajectory of the shortest length, and drilling ricks reduction implies selection of trajectory with minimal"complexity" ("complexity" of the trajectory is discussed, for example, in [5, 6]). One of the significantproblems leading to an increase in the cost of well construction is associated with a loss of wellbore stabilityfollowing with drilling complications, up to tool lost in hole incidents, stuck pipe, stuck casing, need toredrill.

SPE-187830-MS 7

Figure 8—Wellbore stability model

Figure 7 shows a wellbore stability model for deviated well with horizontal completion. Terrigene rockformations 6-17 are characterized by alterating weak clays and argillite with sandstone. The inability tocontrol break-outs by the mud weight has repeatedly led to the critical complications indicated above. Anadditional challenge in well planning is the fact that unstable intervals 14 - 16 are located almost above thetarget layer, which leads to inevitable inclination gain and lowering of the wellbore stability.

To reduce drilling risks, well trajectory optimization using 3D geomechanical model is conducted.3D geomechanical model of oilfield implies ability to predict elastic properties and strength of rock

and near-wellbore stress state along the well path. This allows to predict the zones of possible drillingcomplications and to minimize them. 3D geomechanical model building is firstly motivated by the nessecityof taking into account

a. Fault tectonicsb. Lateral inhomogeneity of mechanical propertiesc. Significant changes in reservoir pressure

3D geomechanical model building includes the following steps:

1. Data audit

a. • Collection and analysis of available datab. • Selection offset wellsc. • Analyze the data quality for geomechanical modeling

2. 1D Geomechanical Model building

8 SPE-187830-MS

a. • Building model for offset wellsb. • Define reservoir conditions to a first approximationc. • Quality control (QC) of 1D model using wellbore stability (WBS) analysis

3. Preparing for 3D modelling

a. • Seismic data cube analysisb. • Geological model analysisc. • Tectonical stress analysis

4. 3D modelling

a. • Computation of elastic and rock strength propertiesb. • Stress cube computation

5. QC of 3D model

a. • Calibrating 3D MEM with 1D MEMb. • Calibrating with direct measurement – microseismic data and minifrac tests

Computation of elastic and rock strength properties for 3D geomechanical model is the same as for 3Dgeological model building. 3D Geomechanical modelling of in-situ stress-state process is based on loadingvertical stress and horizontal stress on boundary walls. Vertical stress is calculated based on overburdenweight. The boundary conditions for horizontal stresses are selected in such a way that the stress statecorresponds exactly to the actual state of the geological environment. In addition elastic properties, thestress-state is affected by structural inhomogeneities and fault tectonics. Results of 3D geomechanicalmodelling are compared to 1D and the results of direct measurements.

Figure 9 shows map of average Unconfined Compressive Strength values (UCS) for one of unstablezones. Star denotes where critical drilling complications took place (stuck pipe, stuck casing) and lead toredrilling of well section. Note lowering drilling risks if well intersects current unstable zone in higher UCS.

Figure 9—Map of average Unconfined Compressive Strength values

Important well trajectory optimization part is the analysis of mud weight window vs well and well azimuthfor every well section. Thus, optimal trajectory is estimated from technological limitations: maximumweight on bit and torque, well length, curvature, mud weight, and well bore stability.

Satisfying geological target, drill bit parameters, well construction limitations, mud weight, maximumdogleg severity, weight on bit and torque etc. means that optimal well from wellbore stability perspective

SPE-187830-MS 9

might not be feasible. Nonetheless, all-inclusive analysis for well construction gives way to improvewellbore stability and to lower drilling risks within the existing limitations.

Figure 10 and Figure 11 illustrate wellbore stability improvement by optimizing well inclination changezone by zone.

Figure 10—Well trajectory optimization.

Figure 11—Wellbore stability improvement through trajectory optimization.

The critical complications leading to significant increase in well construction time were analyzedvs intersection with seismic discontinuities. From the 3D seismic was build cube of attribute "chaos".Although these disorders may be associated with artifacts in seismic data, there might be the case thatsome of them indicate faults and mechanical. Using Ant-Tracking search algorithm that connects mentioneddiscontinuities in seismic (Figure 12 a)), 3D objects were formed. 3D discontinuity objects intersectingwells were compared drilling complications. Taking into account wellbore stability predictions, some of thedrilling events correlated to intersecting with these objects

10 SPE-187830-MS

Figure 12—a) seismic discontintinuities intersected by well . Yellow marked points are where slack-offs occured. Redmarked points are where stuck casing took place; b) avoiding seismic discontinuities. Green-colored is the final well design.

Further well planning, keeping in mind the rest optimization criteria, has taken into account avoiding suckdislocations. Figure 12 b) illustrates shifting existed design to avoid collision. Well with corrected trajectorywas successfully drilled and occurring drilling complications were comparable with unstable zones fromwellbore stability predictions.

Multi-lateral and multi-hole wells construction optimizationMulti-lateral and multi-hole wells trajectory optimization is worth special attention. A good example forthis case is the optimization of the trajectory of the «fishbone».

Reservoir properties of the target formation vary in area and section, this is due to the different reservoirfacies. Such reservoir properties affect its filtration-volumetric characteristics.

The necessary condition for the successful project realization is economic efficiency. Analysis of thegeological situation and mechanical properties of the section showed that «fishbone» technology can besuccessfully used to intensify production. The construction of the "fishbone" resembles a fish bone inits structure, so this construction got that name. This technology allows us to establish hydrodynamicconnectivity and increase the efficiency of reserve development by expanding the coverage zone andincreasing the effective radius of the well in heterogeneous reservoirs.

Today drilling of multilaterals is frequently used enhanced oil recovery method. It is worth noting that thepopularity of this approach is only growing. According to the international classification of the non-profitorganization TAML (Technology Advancement for Multi-Laterals) today there are six levels of branchedwells with different levels of complexity. With an increase in the level, the level of complexity of wellconstruction is also increasing. The complexity depends on the different types of joints and on the type ofisolation of the main bore and side branches.

The third level of TAML was selected for the well construction. This level of complexity means thatthe main bore is cased and cemented, the side bore is cased without cementing and there is a mechanicalisolation of the joint.

Analysis of the destruction of the wellbore according to image data from a previously drilled "fishbone"confirmed the presence of borehole breakouts initiated during drilling in the areas of kick-off (Figure 13).

SPE-187830-MS 11

According to image data borehole breakouts in the areas of kick-off reach values of several meters along thewellbore. This information shows that if the mud weigh is incorrectly selected, we can expect significantproblems in drilling in the areas of kick-off, which can even initiate the loss of wellbore stability.

Figure 13—Images for fishbone sections

First of all, the geological cross section differentiated on the basis by logging and core data. Statisticalanalysis of the wellbore AFB1, BFB1-3, DFB1-3 and horizontal barrel CG1 collapse was helpful to establisha tendency to reduce wellbore collapse conditions mud weight dependency. For the second, with the mudweight change it can be possible to determine minimum and maximum collapse magnitudes (Figure 14).This approach based on the statistical data analysis makes it possible to assess the risks at specific intervalsof kick-off identified by logging curves and a certain spatial orientation of the sidetrack.

Figure 14—Drilled "fishbones" and Boreholes statistics.

12 SPE-187830-MS

Near-wellbore model for fishbone kick-off point with stress-state calculations showed that at the point ofkick-off between boreholes there is a zone of concentrated stresses, which are likely to lead due to formationfailure (Figure 15). Therefore if one to evaluate wellbore stability model for kick-off point, one should setlower rock strength to compensate concentration of stresses.

Figure 15—Joint in «fishbone»

Mechanical earth model built for a vertical well was then tested for other wells including horizontal well.Wellbore stability model demonstrates wellbore failure within highlighted intervals (figures 16-18).

Figure 16—Breakout pressure for horizontal wells, located at an angle of 45 ° to the maximumhorizontal stress, without sidetracking (yellow zone) and with sidetracking (orange zone)

SPE-187830-MS 13

Figure 17—Wellbore stability model in the area of pilot well.

14 SPE-187830-MS

Figure 18—Geomechanical evaluation for "good," "average" and "bad" kick-off intervals.

Thus, numerical calculations of near-wellbore stress-state for the kick-off point helped to imply workflowfor a quick evaluation of optimal intervals for a safe kick-off.

SPE-187830-MS 15

ConclusionWhile the reservoirs difficulties increase and development of construction and completion technologies thegeomechanics became more popular and finds application for solving a wide range of problems.

In order to minimize risks and to reduce NPT of drilling and during the wellbore construction, acomprehensive approach was used. In this case, the synergistic effect from the results of constructing a 3Dgeomechanical modelling and numerical simulation of the stress-strain state of the near wellbore area.

An essential advantage of complementing the stability analysis with a detailed finite-element modelling isthe possibility of taking into account the inhomogeneity of the host environment and the complex geometryof the well. In the case of the analysis of heterogeneities of the medium – bedding planes, faults - the modelmakes it possible to isolate the most hazardous objects for drilling and drilling modes. In the case of awell with branches, this method allows to determine favorable intervals for sidetracking and to optimizethe "fishbone" trajectory.

References1. Ovcharenko Y. Lukin S. Tatur O. Kalinin O. Kolesnikov D. Esipov S. … Podberezny, M. (2016,

October 24). Experience in 3D Geomechanical Modeling, Based on One of the West SiberiaOilfield. Society of Petroleum Engineers. doi:10.2118/182031-MS

2. Konstantinovskaya E. Laskin P. Eremeev D. Pashkov A. Semkin A. Karpfinger F. … Trubienko,O. (2016, October 24). Shale Stability When Drilling Deviated Wells: Geomechanical Modelingof Bedding Plane Weakness, Field X, Russian Platform. Society of Petroleum Engineers.doi:10.2118/182022-MS

3 Kirsch, 1898, Die Theorie der Elastizitat und die Bedurfnisse der Festigkeitslehre. Zeitschrift desVereines deutscher Ingenieure, 42, 797–807.

4 Stefanov Yu. P. Numerical Modeling of Deformation and Failure of Sandstone Specimens.Journal of Mining Science, Vol. 44, No. 1, 2008

5. Oag, A. W., & Williams, M. (2000, January 1). The Directional Difficulty Index - A NewApproach to Performance Benchmarking. Society of Petroleum Engineers. doi:10.2118/59196-MS

6. Samuel R. & Liu, X. (2009, January 1). Wellbore Drilling Indices, Tortuosity, Torsion, andEnergy: What Do They Have To Do With Wellpath Design? Society of Petroleum Engineers.doi:10.2118/124710-MS

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具