WELLBORE INTERVAL DENSITIES
A method for estimating one or more interval densities in a subterranean wellbore includes acquiring first and second axially spaced pressure measurements in the wellbore. The pressure measurements may then be processed to obtain an interval density of drilling fluid between the measurement locations. A tool string including a large number of axially spaced pressure sensors (e.g., four or more or even six or more) electronically coupled with a surface processor via wired drill pipe may be used to obtain a plurality of interval densities corresponding to various wellbore intervals.
This application claims the benefit of U.S. Provisional Application Ser. No. 61/527,948 entitled Interpretation Methodologies and Calculations for Wired Drill Pipe Along String Measurements of Pressure and Temperature, filed Aug. 26, 2011.FIELD OF THE INVENTION
Disclosed embodiments relate generally to geotechnical field measurements and more particularly to Along String Measurements (ASM) that may be incorporated in repeater hardware sections of Wired Drill Pipe (WDP). Methods are disclosed for computing sequential and non-sequential pressure and temperature measurements in these repeaters as well as pressures and temperatures measured by Bottom Hole Assembly (BHA) components. Methods are further disclosed for utilizing these measurements to characterize the subterranean formations, the drilling fluid, and the drilling process.BACKGROUND INFORMATION
During drilling operations, measurements of downhole conditions taken while drilling can provide valuable information that may be used to by a drilling operator to improve efficiency and performance and minimize risk. Such measurements, when transmitted to the surface while drilling, may also provide an essentially real time view of changing downhole conditions allowing for essentially real time performance improvements and risk avoidance. There is considerable interest in the industry in risk avoidance since even relatively minor interruptions in drilling operations can be prohibitively expensive.
The recent introduction of Wired Drill Pipe (WDP) has significantly increased the communication bandwidth between downhole measurement sensors and the surface and therefore the total quantity of data that may be transmitted to the surface during a drilling operation. For example, measurement while drilling (MWD) and logging while drilling (LWD) data, including borehole imaging data, may be readily transmitted to the surface while drilling using WDP. Along string measurements (ASM), for example, including along string pressure and temperature measurements may also be transmitted to the surface during drilling.
While along string pressure and temperature measurements are known in the art, there has been no disclosure of methods for computing sequential and non-sequential pressure and temperature interval densities nor any methods of utilizing such interval densities to characterize the subterranean formations, the drilling fluid, or the drilling process. There remains a need in the art for further development.SUMMARY
Methods for estimating one or more interval densities in a subterranean wellbore are disclosed. For example, a tool string including at least first and second axially spaced pressure sensors may be deployed in a subterranean borehole. Pressure measurements may then be used to compute an interval density between the pressure sensors (i.e., between first and second measured depths in the borehole). The pressure sensors may be internal and/or annular. The tool string may further include a large number of longitudinally spaced pressure sensors (e.g., four or more or even six or more) electronically coupled with a surface processor via wired drill pipe. In such embodiments, a plurality of interval densities may be computed, for example, including a first interval density between first and second measured depths, a second interval density between second and third measured depths, a third interval density between first and third measured depths, a fourth interval density between third and fourth measured depths, and so on.
The disclosed embodiments may provide various technical advantages. For example, the computed interval densities may be further be evaluated during drilling (or other downhole operations) to identify various adverse conditions. These adverse conditions may include, for example, inflow from a subterranean formation to the wellbore, outflow from the wellbore to a subterranean formation, borehole washout, borehole pack-off, poor cuttings transport, and tar intrusion into the wellbore. Early identification of such adverse conditions advantageously enables mitigating actions to be initiated, thereby increasing the likelihood that the adverse condition may be quickly remedied.
In one non-limiting embodiment, a method for estimating an interval density in a subterranean wellbore is disclosed. The method includes (a) deploying a tool string in the wellbore, the tool string including at least first and second subsurface longitudinally spaced pressure sensors deployed at corresponding first and second measured depths in the wellbore; (b) causing the first and second pressure sensors to acquire first and second drilling fluid pressure measurements at the first and second measured depths; and (c) causing a processor to process the first and second pressure measurements to compute an interval density between the first and second measured depths in the wellbore.
In a second non-limiting embodiment, a method for drilling a subterranean wellbore is disclosed. The method includes (a) deploying a drill string in the wellbore, the drill string including at least first and second subsurface longitudinally spaced annular pressure sensors deployed at corresponding first and second measured depths in the wellbore; (b) causing the first and second annular pressure sensors to acquire first and second annular pressure measurements at the first and second measured depths; (c) processing the first and second annular pressure measurements to compute an annular interval density between the first and second measured depths; and (d) evaluating the annular interval density computed in (c) as an indicator of an adverse drilling condition.
In a third non-limiting embodiment, a method for drilling a subterranean wellbore is disclosed. The method includes (a) deploying a drill string in the wellbore, the drill string including at least first and second subsurface longitudinally spaced annular pressure sensors deployed at corresponding first and second measured depths in the wellbore; (b) circulating drilling fluid through the drill string; (c) measuring an annular interval density between the first and second pressure sensors; (d) comparing the annular interval density measured in (c) with a modeled annular interval density to obtain a difference; and (e) evaluating the difference obtained in (d) as an indicator of an adverse drilling condition.
This summary is provided to introduce a selection of concepts that are further described below in the detailed description. This summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used as an aid in limiting the scope of the claimed subject matter.
For a more complete understanding of the disclosed subject matter, and advantages thereof, reference is now made to the following descriptions taken in conjunction with the accompanying drawings, in which:
Drill string 30 includes a plurality of longitudinally spaced wired drill pipe repeater subs 34, at least some of which include annular pressure and temperature sensors 36 and 38. These sensor containing repeater subs may be referred to herein as XLINKS and may optionally further include internal pressure and temperature sensors (not shown). It will be understood that internal sensors are configured to measure the pressure and temperature of the drilling fluid in the drill string 30 while the annular (or external) sensors are configured to measure the pressure and temperature of the drilling fluid in the annulus between the drill string 30 and the borehole wall. Internal and annular pressure and temperature sensors may also be deployed within the various MWD and/or LWD tools included in the BHA 50. Example BHA pressure and temperature sensors are depicted at 52 and 54. The aforementioned pressure and temperature sensors may be in communication with the surface via the high bandwidth digital communications channel such that the along string pressure and temperature measurements may be transmitted to the surface while drilling. The pressure and temperature sensors (or the repeater subs 34) may also include onboard memory for saving the pressure and temperature measurements for later analysis. Other drill-string components (although not explicitly depicted) may also contain annular and internal pressure and temperature sensors, for example, including EMAG repeaters, mud pulse signal boosters and, acoustic telemetry boosters. Pressure and temperature measurements obtained via these sensors may also be transmitted to the surface while drilling (or stored in downhole memory) and utilized in the method embodiments disclosed hereinbelow.
The pressure and temperature sensors may have substantially any longitudinal spacing along the length of the drill string 30. For example, the spaced pressure and temperature sensors may have a longitudinal spacing in a range from about 500 to about 5000 feet in measured depth. Moreover, the spacing between the pressure and temperature sensors is not necessarily uniform. For example, a longitudinal spacing between first and second sensors is not necessarily equal to the spacing between second and third sensors. The disclosed embodiments are not limited in these regards.
The disclosed embodiments are also not limited to the use any particular type of BHA and/or repeater sub pressure sensors. Substantially any suitable pressure sensors may be utilized provided that they provide sufficient accuracy and precision and are robust in demanding downhole environments. For example, pressure sensors that make use of strain gauges (such as those that are commercially available from Paine Electronics, LLC) may be utilized. Likewise, silicon-on-insulator solid state pressure gauges may also be utilized.
It will be understood that the deployment illustrated on
It will be further understood that disclosed embodiments are not limited to use with a semisubmersible platform 12 as illustrated on
The foregoing detailed description is divided into two principle sections, the first describing methodologies for computing interval gradients for along string pressure and temperature measurements. The second section describes methodologies for utilizing the computed interval gradients to interpret various formation and drilling fluid properties and the overall drilling process.Interval Density Computation Methodologies
The density of a fluid under static conditions within the interval between two pressure measurements may be computed from knowledge of a vertical spacing between the pressure sensors and the actual pressure measurements. A temperature gradient can likewise be computed. In general, given a number n spaced apart pressure measurements, a corresponding number of intervals between all sensor combinations (neighbor and otherwise) may be computed, for example, as follows:
Number of Intevals=Σi=1i=n−1(n−i) Equation 1
For example, given 2 spaced apart sensors, 1 interval is available; given 3 spaced apart sensors, 3 intervals are available; given 4 spaced apart sensors, 6 intervals are available, given 5 spaced apart sensors, 10 intervals are available, and so on. In certain of the disclosed method embodiments the number of interval densities computed N may, for example, be in the range: n−1≦N≦Σi−1i=n−1(n−i).
Utilizing any one annular pressure measurement, a density of a fluid (e.g., drilling fluid) under static conditions in a wellbore may be computed, for example, as follows:
where the annular density represents an average density of the annular fluid (e.g., in pounds per gallon), P represents the annular pressure (e.g., in psia), Zmd represents the measured depth of the well, TVD represents the true vertical depth of the well, Inc represents the average borehole inclination, and C1 represents a units conversion constant (e.g., 19.25 ppg/psi/ft).
It will be understood by those of ordinary skill in the art that the density of a fluid may be expressed in various units. The common oilfield unit of pounds per gallon is given in Equation 2. Equivalent vertical head may be used to express the pressure in terms of the vertical height of a column of fluid and may be computed as follows:
where, as is known to those of ordinary skill in the art, vertical head refers to hydraulic head (e.g., in units of feet).Measured Annulus Interval Circulating Density
Of particular interest in this disclosure are methods for computing interval densities (i.e., the density of the fluid) between various spaced apart sensors (e.g., between first and second sensors or between first, second, and third sensors). Utilizing the pressure measurements associated with the endpoints of a specific interval, the density of a fluid between the two sensors may be computed for various specific cases according to the following methodologies. For example, the interval density of a circulating fluid may be computed as follows:
where MA_ICD represents an averaged measured annulus interval circulating density, ΔP represents a change in pressure between first and second measured depths, ΔTVD represents a change in true vertical depth between the first and second measured depths, Pn and Pn+1 represent annular pressure measurements at the first and second depths n and n+1, ZMD(n) and ZMD(n+1) represent the first and second measured depths, and ZTVD(n) and ZTVD(n+1) represent the true vertical depths of the first and second measured depths. Those of ordinary skill in the art will readily appreciate that the true vertical depth (or a change in true vertical depth) may be represented by the measured depth (or a change in measured depth) times the cosine of the average wellbore inclination within an interval.
Under dynamic conditions, e.g., when circulating drilling fluid during a drilling operation, MA_ICD includes the effects of temperature on the compressibility of the input drilling fluid, absolute pressure effects on the density, the volume and mass of the suspended cuttings, the inflow or outflow of drilling fluid between the sensors, and the frictional pressure losses of the circulating mud. This computed interval density (MA_ICD) is described in more detail below via various plots and comparisons with other computed interval densities (e.g., in
Interval densities may also be computed during non-circulating (static) conditions as well using Equation 4. Such conditions are generally available at every connection while adding a pipe stand or a joint to the drill string and occasionally while drilling is suspended during the drilling of a stand. Under such static conditions, the annular frictional pressure losses are absent and the only effects on the interval densities are pressure, temperature, and suspended cuttings effects. This parameter is referred to as MA_ISD and is computed using Equation 4 but under static, non-circulating conditions.
A interval static density may also be computed by subtracting modeled or measured frictional pressure losses from MA_ICD as computed in Equation 4 when computed under circulating conditions. This approach enables a substantially continuous determination of the interval static density and is referred to as MA_ISDmf. Equation 4 may be modified to include these frictional pressure terms as shown below in Equation 5.
Two methods for computing the frictional pressure loss are disclosed; a hydraulically modeled method and an in-situ measurement method. The hydraulic model makes use of various known or estimated fluid and bore properties to compute the frictional pressure loss. The properties may include, for example, temperature, pressure, compressibility, viscosity, flow rate, and flow regime of the drilling fluid, the annular volume of the borehole, the borehole diameter and shape, rotation rate effects, and properties of the borehole wall such as smoothness.
The measurement method may compute the interval density, for example, using Equation 4 under non-pumping static conditions for distinct hole sections or intervals in the well as a function of time. After the pumps are turned back on and before drilling resumes this quantity may be used in the left hand side of Equation 5 along with the measured pressures to compute Pf
In practice it may be advantageous to make use of both the theoretical and measurement methodologies for computing the frictional pressure losses. For example, when the two methods give similar values, the hydraulic model may be used with increased confidence. Differences between the measured and modeled frictional losses may also be used to calibrate the hydraulic model, compute a cuttings density, or flag certain drilling events of interest as described in more detail below.
Upon determining the frictional pressure losses, the measured annulus interval static density MA_ISDmf may be determined while circulating and drilling by substituting the frictional pressure losses into Equation 5. The MA_ISDmf may be computed at various time intervals during drilling.
It should be understood that in drilling operations in which back pressure is applied to the annular fluid (e.g., as is done during managed pressure drilling (MPD) applications), Equations 4 and 5 do not require a back pressure term since a differential pressure is used to determine the interval density. It should also be understood that the interval gradients are a direct function of a down-hole pressure and depth measurements. Therefore any of the principles applied to the interval gradient computations apply to pressure measurements, whether measured or theoretical.Density of Constituent Mud Components
The measured annulus interval static density MA_ISD or computed MA_ISDmf may be taken to be the sum of the individual densities of the individual components of the static annular fluid which may be valid for non-soluble components such as liquid formation fluids and formation cuttings normally encountered during drilling. This may be expressed mathematically, for example, as follows and may enable individual component specific gravities to be computed when their volumetric percentages are known:
where MA_ISDavg represents an average measured annular interval static density, Mi represents the mass of non-soluble component i, and Vi represents the volume of non-soluble component i. MA_ISDavg may also be expressed as a volume weighted average of the individual constituents in the drilling fluid mud. It should be noted that the product of volume and density also represents the mass and may therefore be re-written in terms of volumetric percentages as follows:
where MA_ISDmixture represents the measured annular interval static density of a mixture, Vi represents the volume of non-soluble component i, Vmixture represents the total volume of the mixture, and SGi represents the density (or specific gravity) of component i.
The drilling fluid flowing towards the surface in the annulus generally includes a combination of the drilling fluid that is pumped downward through the interior of the drill pipe and cuttings removed by the drill bit during drilling. The volumetric flow rate in the annulus may be expressed as a combination of these two expected constituents plus an additional term that quantifies increased or reduced flow owing to the addition of an unexpected or unwanted constituent or the loss of a constituent. The additional term may quantify, for example, an inflow of formation fluid into the annulus or an outflow of drilling fluid into the formation. The inflow or outflow may involve either previously drilled or currently drilled formations. Alternatively, the additional term may quantify additional cuttings spalling off the borehole wall after drilling. ASM and corresponding interval density computations may enable the enable these inflow or outflow constituents to be identified and located along the length of the borehole.
As stated above, the annular drilling fluid includes a combination of the drilling fluid that is pumped downward through the interior of the drill pipe and cuttings removed by the drill bit. The cuttings volume may be accounted for by integrating the flow rate in a unit volume of annular fluid over a specified time interval and recognizing that the flow rate out of the unit volume must equal the flow rate into the unit volume. In other words, the flow rate of the mixture may be set equal to the sum of the individual flow rates into this volume. The accumulated volume of the mixture flowing out of the unit annular volume over a given time period may be expressed mathematically, for example, as follows:
where Qmixture represents the volumetric flow rate of the mixture at time t, Qout represents the volumetric flow rate out of the unit annular volume, Qmod in represents the volumetric flow rate of drilling fluid (mud) pumped into the unit annular volume at time t, Qcuttings represents the volumetric flow rate of cuttings flowing into the unit annular volume at time t, and Qx represents the volumetric flow rate of component x flowing in or out of the unit annular volume at time t. Qmud in and Qcuttings may be further defined, for example, as follows:
where TFLO represents the drilling fluid flow rate in units of gallons per minute. TFLO may be determined at the surface using methods known to those of ordinary skill in the art, for example, using the rig pump stroke rate, number of pump cylinders in use, their displacement/stroke, and the pump efficiency. When pumping a compressible fluid such as synthetic oil-based mud (SOBM), the down-hole flow rates tend to change due to pressure and temperature effects on the fluid properties. The measured ASM pressures and temperatures of the interior drill pipe fluid properties may be used to measure the fluid temperature and density in the drill pipe in order to determine the in-situ fluid compressibility and from this calculate the actual down-hole flow rate given the surface flow rate. The downhole flow rate may also be measured downhole.
The volume rate of cuttings being created and flowing into the annulus during the drilling operation may be considered an input variable and may be expressed mathematically, for example, as follows:
where r represents the borehole radius, ROP represents the drilling rate of penetration, K represents percentage of formation porosity destroyed by the crushing action of the bit, and φ represents the formation effective porosity.
The percentage of formation porosity destroyed by the action of the bit K may be estimated by observing the size of the cuttings while drilling. When K is set to unity, the crushing action of the bit destroys all of the porosity, creating cuttings akin to individual sand grains. For example, in unconsolidated sands, the cuttings size will be small and few present with predominantly individual sand grains seen in the samples caught coming from the shale shakers. In shale formations, competent or cemented rock, K is typically less than unity due to the crushing component of the bit being reduced (or minimized depending upon the hardness of the formation).
Determining a value of K may be advantageous in certain drilling operations, for example, when a driller desires to compute an expected volumetric flow rate of cuttings in certain cuttings management programs that determine the volume of cuttings that remain in the borehole and may potentially restrict the movement of the BHA. However, in certain applications it may be sufficient to set K to unity so as to have Qcuttings represent the matrix or rock volume of the formation. This allows the density of the fluid contained within the pore volume to be accounted separately in Equation 11.2 as described in more detail below.
The formation porosity φ may be estimated, for example, from a normalized rate of penetration (ROP) as disclosed in U.S. Pat. No. 4,949,575 or in Rasmus and Stephens (SPE Paper 20443, Real-Time Pore-Pressure Evaluation From MWD/LWD Measurements and Drilling-Derived Formation Strength). However, a fractional volume of fine grained clay/shale/silt in the formation, Vshale, is generally required for this determination. Vshale is normally computed from LWD measurements such as natural gamma ray measurements, however, such LWD measurements are not generally available at the bit.
In certain applications, a dimensionless torque (TD), obtained, for example, from a Mechanical Efficiency Log may be used to differentiate between drilling a porous formation and a shale formation due to the unique and increased dimensionless torque signature of a porous formation as compared to shale. Such differentiation can commonly be made regardless of drill bit type. One example of a Mechanical Efficiency Log is given in Equation 11. Vshale may be estimated from TD and a dimensionless rate of penetration (RD) by realizing that both TD and RD are functions of clay volumes and effective porosity regardless of the wear conditions of the bit (see Burgess, Falconer, and Sheppard, “Separating Bit and Lithology Effects From Drilling Mechanics Data”, SPE 17191, 1988). Such Vshale measurements may then be updated once LWD data above the bit measures the formation properties. TD and RD may be expressed mathematically, for example, as follows:
Where DTOR represents a downhole or surface measured torque, DWOB represents a downhole or surface measured weight on bit, and BS represents a drill bit diameter.
Where ROP represents a rate of penetration and RPM represents a rotation rate of the drill string in revolutions per minute.
The pore fluid contained within the pore space of the formation may be retained within the cutting chip or released into the annular fluid depending on the crushing factor, K. Regardless of the degree of crushing, it will affect the measured interval densities of the annular fluid and may therefore be accounted for separately.
The drilling fluid (mud) flow rate exiting the annulus at the surface, Qmixture or Qout, may also be considered an input measureable volume and may be measured, for example, by a paddle-type measurement placed into the flow out line or by a venturi-type measurement or other means when utilizing managed pressure drilling (MPD) type equipment. This leaves the quantity Qx as the only unknown in Equation 8. In drilling operations this represents one way of detecting a formation fluid inflow or a “kick” (as it is referred) in the industry. However, under conditions in which Qx has been verified to be approximately equal to zero (e.g., via stopping the mud pumps and performing a flow check), Equation 8 may alternatively be used to measure the volume of cuttings flowing into the annulus.
However, in certain applications it can be difficult to utilize the above described methodology to determine Qx given measurements of Qcuttings, Qmud
However, it is desirable to not only know the volume of cuttings being generated, but the density of the cuttings in the annulus between any two ASM pressure measurements since this gives us information as to the type of formation being drilled. Within any two or more arbitrary depths in the annulus, the relative volumetric percentage of the cuttings volume in the annulus makes up a larger percentage than that computed by Equation 8 due to the cuttings travelling upward through the annulus at a lower velocity than that of the drilling fluid. A corrected cuttings volume may be computed by considering a “slip” velocity for the cuttings where Vslip=Vannular−Vcuttings. A transport efficiency FT
where fcuttings represents the volumetric fraction of cuttings in the mud flowing in the annulus, Areaannulus represents the cross sectional area of the annulus a particular depth Z, Qmud represents the volume flow rate of mud from Equation 9, Qcuttings represents the volume flow rate of cuttings from Equation 10, Qpore
The transport efficiency can be computed from empirical correlations such as those disclosed in (i) Sifferman, et al., “Drill Cutting Transport in Full-Scale Vertical Annuli,” J. Pet. Tech., November 1974, 1295-1302; (ii) Moore, “Drilling Practices Manual,” Petroleum Publishing Co., Tulsa, 1974, and (iii) Sample and Bourgoyne, “Development of Improved Laboratory and Field Procedures for Determining the Carrying Capacity of Drilling Fluids,” SPE 7497, 1978. The volumetric fraction of cuttings flowing in the annulus is also a function of wellbore inclination since the cuttings tend to fall out of suspension in high inclination sections. The constant a is used to account for the fact that as the wellbore becomes closer to horizontal, the cuttings tend to drop out of suspension and are transported along the wellbore in a “saltation” type mechanism. The inclination and saltation terms in Equation 12 are intended to result in a net upward or vertical cuttings slip velocity. Equation 12 may then be rearranged to compute the term fcuttings, for example, as given in Equation 13.
where X=cos Inc+a sin Inc.
Being liquid at downhole temperatures and pressures, the formation pore fluid volume that is released into the annulus may have negligible slip velocity with respect to the mud. The fractional volume of the pore fluid fpore
In some applications, especially at shallower depths, the formation pore fluid volume fpore
Transformation from volumetric or fractional flow dimensions to a depth dimension requires the simultaneous consideration of cross-sectional areas and fractional volumes. The annular volume may be represented mathematically, for example, as follows:
where Volannulus represents the annular volume between any two depths z=n and z=n+1, Dbh represents the borehole diameter obtained for example from the bit diameter or LWD caliper measurements, and D P represents the diameter of the drill pipe located between z=n and z=n+1. Equation 14 assumes a borehole having a circular cross section. This assumption may be suitable for many drilling operations, however, the disclosed embodiments are not limited in this regard. For example, a more general elliptical shape may be utilized.
It will be understood that Equation 14 is expressed in terms of borehole depth rather than time. It will further be understood that the link between the volumes and depth is the annular velocity of the mud and cuttings mixture, while the link between the depth based annular volume and time is the rate penetration. Thus the annular volumes and fluid flow rates may be expressed alternatively as functions of time or depth. For example, the cuttings and fluid flow velocity may be integrated over a specific time period to determine the cuttings as a function of depth.
In one workflow example, an array of annular volume over discrete depth intervals may be computed using Equation 14. The array may be as fine as a few inches in depth or as sparse as one to two feet in depth. In the lower BHA (below an LWD caliper tool), the bit size may be used as the borehole diameter. The diameter may be updated using measured values when LWD caliper measurements become available at the predefined depths. The diameter of the drill pipe may also be continually updated using discrete functions of time as the various pipe diameters pass through these same depth points and the various cuttings are lifted from the bit face and carried into the annular volume. The terms Qmud in and Qcuttings may be computed from Equations 9 and 10 at discrete time intervals (e.g., every few seconds). These volumes may then be utilized in Equation 13 to compute the fractional volume of cuttings within each discrete time period. The velocity of the cuttings may be integrated to obtain the corresponding depth position of the cuttings with time and may be expressed mathematically, for example, as follows:
It may improve accuracy to integrate the mud annular velocity (as opposed to or in addition to the cuttings velocity) due to the higher fractional volume and larger volumetric flow rates. This may be expressed mathematically, for example, as follows:
Equations 15 and/or 16 may be used to generate multi-dimensional arrays indexed by depth increments. Each column represents one chosen time interval and may contain TIME, as well as Areaannulus, Qmud in, Qcuttings, Qpore
The quantities MA_ISD and MA_ICD described above and calculated using the ASM data and Equation 5 may include multiple depth intervals within the previously described depth array. This multi-dimensional array may be integrated over the depth intervals corresponding to the ASM interval to derive an averaged density of the mixture which may be compared directly with the ASM measured values. A similar process may also be followed for the fractional cuttings volume. From Equation 7, MA_ISDmixture may be expressed mathematically, for example, as follows:
where fcuttings, fpore
Equation 17 may be further expanded by considering the pore fluid to include a combination of hydrocarbons and water that may or may not have been flushed by the drilling mud. The expanded form of Equation 17 may be represented mathematically, for example, as follows:
where F represents a flushing factor such that 1≦F≦0 with F=1 representing no flushing and F=0 representing complete flushing, Sw represents pore water saturation, 1−Sw represents pore hydrocarbon saturation, SGpore
When drilling under conditions of no influx or outflux (i.e., fx=0), Equation 17.1 includes four unknowns (SGcuttings, F, Sw, and SGpore
Given that Equations 17.1 and 17.2 include at least four unknowns, various techniques may be utilized to determine which water saturation is appropriate. For example, by assuming no flushing (F=1), inputting SGcuttings from the known lithology (e.g., shale or porous formation as determined by MEL), and assuming a value for SGpore
Computing Sw requires that the hydrocarbon density be input. Since this quantity is unknown, Sw may be computed based on a first hydrocarbon density representing gas (SGgas≈0.2) and a second hydrocarbon density representing oil (SGoil≈0.8). When the formation is gas bearing, the computed Sw using SGoil is typically less than zero and therefore erroneous. When the formation is oil bearing, the the computed Sw using SGgas is typically between zero and one, but erroneously high. The computed Sw using SGgas advantageously represents an upper bound on the actual water saturation.
When inflow is detected, the quantity fcuttingsSGcuttings may be assumed to be constant for a time interval. Equation 17 may then be used to compute fxSGx from which SGx may be computed when fx is known (e.g., from Equation 8). Determining (or estimating) SGx can be advantageous in determining the type of fluid inflow into the wellbore.Measured Drill Pipe Internal Interval Static and Circulating Density
The aforementioned internal ASM pressure sensors that are deployed and configured to measure an internal pressure of the drill pipe (ASMinternalP) may be used to obtain internal fluid gradients within the drill pipe under no flow (MIF_ISD) and flowing conditions (MIF_ICD), for example, using Equation 4. The difference between MIF_ISD and MIF_ICD is generally due to frictional losses in the drill pipe. When two axially spaced pressure sensors are sufficiently close to the bit and separated in TVD so as to give adequately high signal/noise measurements, the internal interval static density can be measured when not pumping. The internal interval static density may also be computed using Equations 4 and 5 as described above to determine the frictional pressure losses and to subtract them from the measured internal dynamic interval density. Frictional losses may also be computed using a hydraulics model.
The measured internal interval static density (MIF_ISD) is a function of the density of the actual fluid being pumped into the pipe at the surface plus any pressure and temperature effects that affect the compressibility of the fluid. If the sensor pairs are far above the bit, a computed temperature correction to the interval static density may be applied using an appropriate hydraulics model that includes temperature and frictional pressure effects.
MIF_ISD represents the fluid exiting the bit before any cuttings loading and annular frictional loss effects and may therefore be used as the input to the computation of the expected annulus fluid interval static density described in more detail hereinbelow.Expected Drill Pipe Internal Interval Static and Circulating Density
Known hydraulic modeling techniques may be utilized to predict the internal fluid density as a function of the internal (predicted or measured) pressure and temperature using the surface mud density properties as a base fluid for the modeling. The surface mud properties are typically measured by mud loggers but may also be measured by sensors at the surface. Accounting for the pressure and temperature effects results in an expected internal fluid interval static density EIF_ISD. By taking into account modeled frictional effects an expected internal fluid interval circulating density EIF_ICD may be obtained. Expected interval densities are also referred to herein as modeled interval densities. The expected internal densities are generally equal to the measured quantities MIF_ISD and MIF_ICD when the hydraulic model is correct. A minimization process may be used to adjust appropriate hydraulic parameters until a suitably accurate match is found.Expected Annulus Fluid Interval Static Density
An expected annulus fluid interval static density (EAF_ISD) may be obtained by correcting MIF_ISD for pressure and temperature effects as the input mud flows up the annulus to the surface. The EAF_ISD may be compared to the various measured interval densities to identify certain undesirable drilling events as described more detail below in various applications of the INTERVAL DENSITY APPLICATIONS section of this disclosure. The annulus pressure and temperature are typically measured by the ASM sensors in the WDP. When these measurements are not available, and only the BHA sensors are present, pressure and temperature gradients may be assumed between the BHA sensors and the surface.Expected Annulus Interval Static Density
The fluid leaving the bit and being pumped into the annulus is a fluid having properties defined by EAF_ISD, which as is described above is MIF _ISD corrected for pressure and temperature effects on the density. The cuttings load (with Qx=0) computed using one or more of Equations 8-16 may be added to EAF_ISD to obtain an expected annulus interval static density EA_ISD. Expected interval densities are also referred to herein as ‘modeled’ interval densities. The EA_ISD represents a hypothetical fluid having the properties of the mud being injected into the annulus at the bit loaded with the drilled and suspended cuttings having a certain interval density and may be expressed mathematically, for example, as follows:
The difference between EAF_SD and EA_ISD is due to the cuttings loading. If the difference is minimal at the bottom of the hole, the cuttings density and loading effects computed using Equations 8-16 is likely correct. Given a discrepancy, the cutting density may be adjusted. If MA_ISD decreases and drops below EA_ISD as the mud flows up the annulus into the deviated section of the borehole, it indicates that the cuttings may be dropping out of suspension and settling at the bottom of the borehole. Moreover, inflow or outflow from the wellbore may result in differences between these two computed parameters and may be used to flag lost circulation and gas kicks.Expected Annulus Interval Circulating Density
Taking the computation of EA_ISD and adding the annular friction pressure losses to it results in an expected annulus interval circulating density EA_ICD. This parameter is a function of the input mud density adjusted for temperature, pressure, cuttings load, and annular frictional pressure losses and is therefore comparable to MA_ICD. The expected and measured quantities (EA_ICD and MA_ICD) tend to be equal to one another when the cuttings density and the frictional losses are input correctly. When these quantities are not equal (or not close to equal), it may indicate a change in cuttings density from the assumed cuttings density or inflow or outflow event (a Qx event). EA_ICD may be expressed mathematically, for example, as follows:
where ZTVD(n) and ZTVD(n+1) represent the true vertical depths of the well at the first and second depths n and n+1 and Pf represents the frictional pressure drop acting on the fluid above the sensor as described above with respect to Equations 4 and 5.Equivalent Top of Fluid Level
The equivalent measured or true vertical depth of the top of the fluid level may be computed from the annular mud interval density existing between any two pressure sensors using the concept of hydraulic head. This may be referred to as the equivalent top of fluid level (ETOFL) and is intended to define the uppermost depth or level that a fluid would occupy if it were continuous and had the same properties as the fluid between the two measured pressures. A back pressure may sometimes be applied to the annular choke during managed pressure drilling (MPD) operations. With an incompressible fluid in the annulus, the pressure may be subtracted from the measured pressure to compute ETOFL. When the fluid is compressible, simply subtracting the back pressure may not to be suitably accurate such that it may be necessary to compute an equivalent back pressure at the sensor. Such calculations may be accomplished, for example, using hydraulic models.
The following mathematical equations may be used to compute ETOFL in the presence of an applied back pressure using the previously calculated interval densities. In these equations, a positive ETOFL indicates that the computed fluid level is below the surface, while a negative ETOFL indicates the fluid level is above the surface.
where ETOFL represents the equivalent top of fluid level which is essentially equivalent to the fluid elevation in a well including a fluid having a static density, P represents the measured pressure, Pf represents the frictional pressure loss, BP represents the surface annular applied back pressure, n represents a pressure sensor at some measured depth, and n+1 represents a pressure sensor at some deeper measured depth.Theoretical or Extrapolated Surface Annular Back Pressure
In MPD operations it may be useful to compute a theoretical or extrapolated surface annular back pressure (BP) from the measured downhole annular pressures and to compare the computed values with the actual surface annular back pressure (SBP). Automated software routines may then be employed to adjust the actual applied BP so as to minimize any differences to maintain a constant bottom hole pressure (BHP).
Equations 20.1 and 20.2 show that an increase in the interval density at a given BP results in an increase in ETOFL. This increase in interval density may cause the theoretical back pressure in Equations 20.1 and 20.2 to decrease and even go negative in some cases. In an event causing a sudden increase in the annular pressure measured by the lowermost pair of sensors (e.g., due to a restriction in the drill string above the sensors), the lowermost interval density remains substantially constant, ETOFL decreases, and the computed surface annular back pressure (SBP) increases. Since the theoretical BP depends on the interval from which it is computed and the impact that various events have on the interval density, interpretation of the theoretical BP is application dependant as described in more detail below with respect to Table 10. In general interpretation of the theoretical BP is used in combination with a computed interval density in order to obtain the proper action for adjusting the actual surface back pressure.
The theoretical back pressure BP may be expressed mathematically, for example, as follows:
where BP represents the theoretical back pressure, Pn and Pn+1 represent the measured pressures at sensors n and n+1, and ZTVD(n) and ZTVD(n+1) represent the true vertical depths of sensors n and n+1.Velocity and Acceleration of Interval Density Changes
It is often desirable to know the direction and degree of change in the computed interval specific gravities with time in order to determine if the system is tending towards stability or instability, and for example, tracking an inflow as it moves up the annulus. The rate of change of the interval density may be represented mathematically, for example, as follows:
where VID represents the rate of change of the interval density with time and IDt represents one of the interval densities described above at time t.
A further derivative of the rate of change (i.e., an acceleration) may also be useful in determining the direction of the change and how quickly the interval density is changing in order to determine the necessary reaction time for remedial action. The acceleration may also help distinguish between gas kicks versus water or oil inflows. Interval density acceleration may be expressed mathematically, for example, as follows:
where AID represents the rate of change of the velocity of the interval density with time (i.e., the rate of change of the rate of change of the interval density) and VIDt represents one of the velocities of the interval densities at time t.Interval Density Applications
In this section methodologies for interpreting the computed interval densities are presented along with several applications for using computed interval densities to determine, diagnose, manage, and/or remedy various drilling events.Interpretive Methodology
Table 1 summarizes the various interval densities described above in the INTERVAL DENSITY COMPUTATION METHODOLOGIES section and the physical effects that are included in each. The mathematical equations listed above may be used to compute the various interval densities. The computations may be performed in substantially real time while the well is being drilled or subsequent to the drilling operation using recorded historical data. The disclosed embodiments are not limited in this regard. The computed interval densities as well as their depth and time relationships may be plotted on various crossplots or other displays enabling the driller (or a computer software program) to recognize, differentiate, and take control of mitigating various situations discussed later in this section. Moreover, use of the computed interval densities is not limited to drilling operations, but may also be useful in various completion and production operations.
EIF_ICD and EIF_ISD are the modeled (expected) internal interval circulating and static densities computed using the surface input mud properties, including downhole pressure and temperature in the drill string at the depth of interest. The expected quantities may be compared directly with the measured internal interval circulating and static densities MIF_ICD and MIF_ISD. MIF_ISD may be obtained by subtracting an internal frictional pressure loss from the measured MIF_ICD or by direct measurement. The frictional pressure losses may be obtained via modeling and/or measurements. For example, MIF_ICD may be measured directly by measuring MA_ISD when the mud pumps are turned off (e.g., when adding a length of drill pipe to the drill string). The difference between MIF_ICD measurements made while circulating and not circulating (when the pumps are on and off) may be considered to be a direct measurement of the internal frictional pressure losses (ΔP— Internalfric).
The modeled EIF_ISD may be compared with MIF_ISD (which is MIF_ICD-ΔP_Internalfric when circulating and MIF_ISD when not circulating). An error minimization process (or a manual procedure) may be used to adjust the hydraulic model parameters that account for pressure and temperature effects on the drilling fluid such that EIF_ISD equals MIF_ISD. A subsequent error minimization process may then be employed to adjust the hydraulic model parameters that account for internal frictional pressure losses such that EIF_ICD equals MIF_ICD (i.e., such that the modeled frictional pressure loss equals to the measured frictional pressure loss ΔP_Internalfric). Iterative minimization processes may be utilized to provide for accurate results. The minimization processes may also be repeated at various flow rates and the results stored in a look-up table for future reference.
The hydraulic model parameters obtained above for the pressure and temperature effects on the input mud properties may be utilized in the annulus environment as well. The annular fluid properties as a function of the annular pressure and temperature may be input to the hydraulic model to obtain a modeled (expected) annular fluid interval static density EAF_ISD. This parameter represents the interval density of the annular fluid (without cuttings and friction effects) as a function of annular pressure and temperature as a function of depth and time. Calibration and determination of the annular friction effects may be accomplished in the same manner as described above for the internal frictional effects. For these minimizations, EA_ISD, EA_ICD, MA_ISD and MA_ICD are computed as opposed to EIF_ISD, EIF_ICD, MIF_ISD and MIF_ISD as described in the preceding paragraph.
The modeled annular interval static density EA_ISD may be utilized as the input mud properties with annular pressure and temperature and modeled cuttings effects included. EA_ISD may be equal to MA_ISD when the generation and transport of cuttings in the annulus is properly modeled and the modeled frictional pressure losses that are subtracted from MA_ICD are correct. An error minimization process may be utilized to compute a cuttings density using appropriate values for frictional transport efficiency, ROP, porosity, and the density of the cuttings free fluid flowing in the annulus determined from the minimization described above for EAF_ISD. Changes in the computed cuttings density by interval may indicate that cuttings are dropping out of suspension since the modeled cuttings density is constant with depth. A cuttings management process may track the loss of cuttings in the annulus and indicate the potential for undesirable drilling events such as pack-offs while drilling or while reaming or pulling out of the hole.
Disclosed method embodiments may further utilize measurements of the actual flow into and out of each interval (e.g., as described above with respect to Equation 8). Such measurements provide for a determination of Qx and may therefore be used to differentiate between inflow or outflow effects versus incorrect cuttings modeling effects such as the assumed cuttings density. When flow in does not equal flow out, differences may be attributed to the quantity fx·SGx in Equation 17 indicating flow in or out of the annulus in the interval in which the difference occurs. In certain applications the interval densities may then be used to compute the fractional volume and density of an inflow material (e.g., using Equations 8-17). This process may be useful for distinguishing between gas and salt water kicks, for example.
MA_ICD and EA_ICD may be equal when the various parameters discussed above are modeled correctly. Differences between these two quantities may also indicate undesirable drilling events as discussed above. Additionally, modeled frictional effects may depend on the borehole diameter. Using an LWD caliper, these effects can be properly accounted for. However, with time the borehole wall may experience washout or enlargement, for example, due to drilling practices, shale stability, or other geomechanical effects. Differences in MA_ICD and EA_ICD may be used to detect and monitor borehole diameter changes. A minimization process may also be used to determine the average borehole size within each interval as a function of time.
The annular frictional losses also depend on the drill pipe rotation speed (RPM) and fluid flow rate. Since these parameters may change with time, the annular frictional effects can also be time dependant and may be accounted for during drilling.Effects of Pressure and Temperature on Fluid Densities
The fluid or mud being pumped into the well while drilling may be affected by the pressure and temperature changes it undergoes as it travels down the drill pipe and back up the annulus. For example, pressure and temperature changes cause corresponding changes to the density of the fluid. These changes may be measured using the aforementioned ASM measurements and may enable the relationship between fluid density and pressure and temperature to be quantified and/or modeled which in turn enables other effects such as cuttings loading and friction to be determined.
Internal ASM pressures, temperatures, and computed interval densities and temperature gradients may be used with a hydraulic model to calibrate the model parameters. The hydraulic model may then be used to predict the effects at any other point in the system as a function of depth and time. Annular measurements may be used in the same manner under non-drilling conditions (i.e., when there are no cuttings in the annular fluid). When the hydraulic model parameters are well defined and predictable for a particular drilling fluid, and in cases where either a measured temperature or measured pressure is not available, the hydraulic model may be used to predict the missing measurement.
Table 3 summarizes the parameters depicted on
With continued reference to
The computed interval densities are also shown in track 4 (506) and are labeled as MA_IED—003—001 (the interval density between the APRS and 1244 sensors), MA_IED—003—009 (the interval density between the 1244 and 1231 sensors), and MA_IED—999—009 (the interval density between the 1244 ASM sensor and the surface annular pressure sensor). When the pumps are shut down at the connection, the interval densities drop due to the elimination of annular friction losses. The interval densities are essentially the aforementioned quantities MA_ICD when circulating and MA_ISD when not circulating. In this particular example, the interval densities also closely represent the EAF_ISD since the rate of penetration (ROP) was low and there were long periods of circulation between drilling events, implying there were little to no cuttings suspended in the annular fluid.
The uppermost interval density (MA_IED—999—009) is approximately equal to the computed equivalent densities shown in track 3 (at 8 ppg). As depicted in track 4, the interval densities decrease significantly with increasing depth, with MA_IED—003—009 being about equal to 7.6 ppg and MA_IED—003—001 being about equal to 7.3 ppg. The decreasing interval densities are likely due to increasing temperatures lower in the wellbore. Absent such temperature effects, one would expect the density of a compressible fluid such as an OBM to increase with increasing depth. However, as shown on
With still further reference to
In well drilling operations, the borehole temperature commonly increases with increasing depth. Under circulating (and drilling) conditions, the temperature of the drilling fluid is generally not a strong function of depth (due to the mixing of the fluid and transport back to the surface). When circulation stops, the temperature typically increases with time and any particular depth until a steady-state temperature is reached. As a result, the density of the drilling fluid may also be expected to decrease with time after circulation ceases. These time dependent changes in density may also be observed using the aforementioned interval densities.
The ASM pressure and temperature measurements and their relationship to fluid density may be further utilized in refining and/or calibrating conventional hydraulic models. For example, the measurements may be utilized to determine the coefficients in the conventional API-13D equations:
ρbase=(a1+b1P+c1P2)+(a2+b2P+c2P2)T Equation 24
ρbrine=(a3+b3P+c3P2)+(a4+b4P+c4P2)T Equation 25
where ρbase represents the density of the base drilling OBM, ρbrine represents the density of the brine, P represents pressure, T represents temperature, and a, b, and c represent fitting coefficients. Table 4 includes sample “book” values for various conventional oil and/or brine solutions as well as fitting statistics and range of validity.
It may be advantageous in certain applications to adjust these “book” values according to in-situ conditions. Since the oil to water ratio is known (it is commonly controlled at the surface), Equations 24 and 25 may be combined into a single equation having six coefficients, for example as follows:
ρmud=(i1+j1P+k1P2)+(i2+j2P+k2P2)T Equation 26
where ρmud represents the density of the drilling fluid (the combination of base and brine) and i, j, and k represent the coefficients. This density may be measured in-situ, for example, using the aforementioned interval density computations where the pressure and temperature values represent an average value for the interval.
A drill string including six ASM pressure and temperature sensors, for example, may enable the six coefficients to be computed. For example, six interval densities may be calculated using the corresponding six pressure and temperature measurements to obtain six equations having six unknowns (the six coefficients). Values for the coefficients may then be determined using conventional root finding algorithms. It will be understood that the necessary number of intervals may be reduced, for example, via using minimization techniques or using interval densities computed at multiple times (or multiple depths) provided that the pressure and temperature measurements are sufficiently different.
Alternatively, Equations 24 and 25 may be combined into a single equation having twelve coefficients, for example as follows:
where VVase and Vbrine represent the volume fractions of base and brine. The coefficients in Equations 27 and 28 may be obtained by making 12 independent interval density measurements, for example, at two distinct locations using the drill string described above having six ASM pressure and temperature sensors.
In another alternative embodiment, values for the brine coefficients (a3, b3, c3 and a4, b4, c4 in Equations 25 and 28) may be assumed and the six base coefficients evaluated, for example, using at least six independent interval density measurements.
In the foregoing embodiments, the coefficients may be determined using either internal interval density measurements or annular interval density measurements. Internal interval density measurements may be preferred due to the lack of cuttings in the interior of the drill string, however, annular measurements may also be utilized when the cuttings are accounted for using one or more of the aforementioned techniques.Cuttings Transport Efficiencies and Formation Characterization
ASM pressure and temperature measurements may be utilized to detect changes in cuttings densities and transport efficiencies and may therefore further be utilized in characterizing the lithology of the formation being drilled. As described above with respect to Equations 8-17, the ASM pressure measurements may be used to determine constituent densities of various materials in the drilling fluid. In operations in which there is no annular inflow or outflow (i.e., when Qx and fx are approximately equal to zero), the cuttings density may be readily determined using EA_ISD and MA_ISD.
Track 1 depicts (at 602) MIF_ISD and EIF_SD, the former of which is computed from MIF_ICD by subtracting the modeled and/or measured internal drill pipe frictional effects on the flowing mud. EIF_ISD represents the input mud density properties corrected for the effects of the internal drill pipe measured and/or modeled pressures and temperatures using a suitable hydraulic modeling program. The necessary hydraulic modeling parameters for the pressure and temperature effects may be determined by matching EIF_ISD to MIF_ISD over the intervals where MIF_ISD computations are available.
Track 3 includes (at 606) the computed annular interval densities, EAF_ISD, MA_ISD, EA_ISD, MA_ICD, and EA_ICD. EAF_ISD represents the density of the cuttings free input mud flowing up the annulus corrected for the measured annulus pressures and temperatures using the same hydraulic modeling parameters determined for the internal mud. The modeled cuttings load is added to EAF_ISD to obtain EA_ISD. The measured interval static density MA_ISD is equal to the measured interval circulating density MA_ICD less the annular frictional losses when the cuttings volume, density, and transport and the frictional flow parameters are properly modeled. A minimization program may be utilized in the modeling as described above in to achieve this as described above.
Track 4 depicts (at 608) the computed cuttings density. Other parameters are shown on Tracks 5-8 and discussed in more below with regards to other examples. It will be understood in
Time differentials of the measured interval static and circulating densities MA_ISD and MA_ICD are shown in track 5 at 610. Equivalent top of fluid ETOFL for the static and circulating fluid are shown in track 6 at 612. Calculated annular back pressure BP for the static and circulating fluid are shown in track 7 at 614 and the measured annulus static and circulating pressures P are shown in track 8 and 616.
It will be understood that a change in cuttings density may be identified by signatures other than those discussed above with respect to
The cuttings density SGcuttings may be used, for example, to identify the lithology of the formation being drilled (e.g., sandstone, limestone, dolomite, shale, tar, salt, etc.). For example, quartz sandstone has a density of about 2.65, calcium carbonate limestone has a density of about 2.71, calcium magnesium carbonate dolomite has a density of about SG of 2.85, mixed mineral shale formations have an average density in the range from about 2.6 to about 2.7, halite salts have a density of about 2.17, tar layers have a density in the range from about 0.8 to about 1.1, and anhydrite has a density of about 2.97. Knowledge of the cuttings velocity (or velocities) with time, enables cuttings depths to be assigned, which in turn may enable a lithology log (e.g., as depicted in track 2) to be constructed. In the example depicted on
Those of ordinary skill in the art will readily appreciate that formation bulk density is a widely used petrophysics parameter. This parameter is commonly used for applications ranging from overburden calculations, geomechanical modeling, synthetic seismograms, and formation porosity determination. The formation bulk density is generally a function of the lithology (or mineral content of the formation) and the fluid type and volume in the formation. In drilling operations in which the drilling process destroys the formation porosity, the computed cuttings density may be used as the mineral density (formation matrix density) to compute the porosity from a borehole geophysical measurement of bulk density.Tar Mat Identification
Tar zones (also referred to in the art as tar mats) are a common threat in drilling operations and can at times represent a serious risk to a drilling operation. Since tar is difficult to identify in seismic maps, avoidance can be challenging and often relies primarily on local experience. Moreover common utilized logging while drilling (LWD) technologies, such as gamma ray and resistivity logging measurements, are not always capable of identifying tar zones. As such a drilling operator sometimes does not realize that a tar zone has been intercepted until the annulus is full of tar. This can result in a pack-off situation and a stuck BHA. The ASM pressure and temperature measurements and the interval densities disclosed herein may be used to quickly identify and mitigate intercepted tar zones.
The disclosed interval densities may be utilized to identify tar in the annulus via computing the interval cuttings density as described above with respect to
Early identification of tar mats enables the drilling operator to mitigate the influx of tar into wellbore. Such mitigation may include any number of techniques, for example, including, the use of managed pressure to artificially boost the constraining pressure or back pressure in the annulus to keep additional tar from sloughing into the borehole, moving the pipe up above the point of the tar mat without circulating, then introducing a heavier weight mud into the borehole (called spotting a pill), side tracking around the tar, treating the tar with various chemical additives, and isolating the tar via the use of various types of casing. The disclosed embodiments are, or course, not limited to any particular mitigating action.Borehole Washout
Due to various geomechanical and/or drilling practices the borehole can become enlarged with time during a drilling operation. Such borehole enlargement can be detrimental for several reasons. For example, an enlarged borehole can reduce the velocity of cuttings moving up through the annulus thereby enhancing the possibility of cuttings dropping out of suspension and packing off the borehole. Enlarged boreholes also require larger volumes of cement during casing operations.
It will be understood that a change in borehole diameter (e.g., caused by a washout) may cause corresponding changes in certain of the disclosed parameters other than those described above with respect to
As used in the art, a pack-off describes a situation in which the borehole diameter has been reduced creating a “choke” to fluid flowing up the annulus. Such a reduction may be caused, for example, by a large volume of cuttings that have dropped out of suspension in the annulus or sloughing of the borehole wall into the annulus. With insufficient annular fluid velocity, mud viscosity, or in a highly inclined borehole, the cuttings may accumulate at some depth in the well and cause a restriction (the pack-off). Depending upon the severity of the pack-off, the pressure may increase to undesirable levels deeper in the well and may even cause the formations to fracture if remedial action is not performed in a timely manner. The pack-off can also result in lost circulation which in turn can cause a loss of hydrostatic head and a possible inflow or even a kick from a permeable formation. A severe pack-off can even also result in a stuck BHA if sufficient cuttings are allowed to accumulate around the drill string.
The pack-off is depicted schematically in track 3 (at 1202) in
It will be understood that the development of a pack-off or a restriction may cause corresponding changes in certain of the disclosed parameters other than those described above with respect to
The identification of the pack-off by observing annular pressures and interval densities may be automated such that the signature shown in
As is known to those of ordinary skill in the art, formation fluids tend to flow into the wellbore during drilling when the formation has a higher pore pressure than the mud pressure at the formation depth. Such inflow events can occur further up the borehole if the mud column is allowed to drop below the surface, fore example, when tripping the drill pipe out of the borehole. Swab events can also contribute to an inflow. Formation fluids, such as gas, oil, or connate water, generally exhibit a lower density than the drilling mud. Any inflow therefore tends to further reduce the hydrostatic head, allowing the inflow rate to increase until the wellbore can no longer be controlled. Timely mitigation therefore requires early recognition of the inflow event. ASM pressure and temperature measurements and the disclosed interval densities may be used to identify inflow events soon after they begin.
With continued reference to
With continued reference to
It will be understood that the development of an inflow (or kick) may cause corresponding changes in certain of the disclosed parameters other than those described above with respect to
During formation fluid sampling operations, formation fluid may be pumped (or released) into the annulus. For example, formation fluid is often pumped into the annulus for a period of time prior to sampling the formation fluid to ensure that only virgin fluid is sampled (i.e., that the sampled fluid is not contaminated with drilling fluid or cuttings). Up to one barrel or more of formation fluid may be released into the annulus for each sample acquired. The density of the annular fluid may be monitored while sampling using the interval density techniques describes herein. Moreover, after the samples are acquired, the formation fluid may be circulated to the surface and released through an annular choke. The interval densities may also be used to monitor the upward movement of the formation fluid through the annulus, thereby potentially saving considerable rig time.
When an inflow event (e.g., a kick) is encountered, a drilling operator may elect to circulate through an annular choke while heavy mud is pumped downhole. The disclosed interval densities may continue to be measured and computed and used to determine when the bottom hole density and pressure is sufficient to stop the inflow. For example, a measured bottom hole pressure may be used to drive a choke to keep the pressure within a desired range while pumping the heavy mud.
Outflow from the Borehole Annulus
Annular fluids may flow into the formation as it is drilled when the formation has a lower pore pressure than the drilling fluid pressure at that depth. Such an outflow may happen at the bit or further up the borehole if the drilling fluid pressure is allowed to increase above the formation pressure. In some operations, an outflow reduces the hydrostatic head thereby causing the outflow rate to decrease until the wellbore stabilizes. Such outflow events may be thought of as self-mitigating. However, in other operations, the reduced hydrostatic head caused by the outflow may trigger an inflow (or kick) in another formation (e.g., at another location in the borehole). As described above, inflow events can lead to highly dangerous and uncontrollable well conditions. Timely mitigation requires early recognition of the problem, and in keeping with the purposes of this section, timely recognition of the outflow event. ASM pressure and temperature measurements and the disclosed interval densities may be used to identify outflow events soon after they begin.
With continued reference to
With continued reference to
It will be understood that while the annular fluid level may drop during a lost circulation event, the internal drill-pipe fluid level may or may not coincide with the annular fluid level due to differing pressures above and below both fluid levels. This condition is sometimes referred to as in the art as “U-tubing”. Internal pressure measurements may be used to determine the fluid levels in the interior of the drill-pipe in an analogous manner to the method described above for the annular fluid level. Moreover, in extreme lost circulation events, the fluid level in the annulus may drop during circulation while drilling fluid is being pumped down the interior of the drill string.
It will be understood that the development of an outflow may cause corresponding changes in certain of the disclosed parameters other than those described above with respect to
In response to an outflow event a drilling operator often shuts in the well, stops pumping, and closes the annular choke until pressures stabilize. The interval densities may be utilized to determine the liquid level of the drilling fluid while the ASM and APWD measurements may be used to obtain the BHP when the liquid level stabilizes. This BHP then becomes the maximum BHP that should be applied during the future drilling operations. When drilling restarts, the flow rate may be reduced and/or nitrogen may be injected into the input flow stream to reduce the density of the drilling fluid sufficiently so that the BHP remains below the maximum value. The average calculated annular BP or any one of the interval calculated BP or the downhole measured annulus pressures may be used in an automatic choke control. As disclosed herein, the choke position may be controlled in time intervals by an electro-mechanical server to reduce the BP by the amount calculated until the system stabilizes.
In the depicted example, downhole dynamics sensors detected a high degree of stick/slip in a measured depth range from about 5152 to about 5179 meters. A viscous pill was pumped on 14-Dec 16:00 one while the back pressure was kept at 350 psi. This was observed to stabilize the whole and drilling continued at a controlled rate of penetration to 5199 meters. On 15 Dec 07:20 the applied torque increased from 8000 to about 12,700 foot pounds and partial fluid losses were thought to occur based on bit level observations. At 07:42 pressures were observed to drop significantly in response to a lost circulation event and a loss of hydrostatic head. At the APRS sensor, the pressure dropped from about 7500 to about 6800 psi as indicated at 2102. The interval density between the APRS and 1244 pressure sensors also dropped from about 8.5 to about 5 ppg as indicated at 2104, while the other two interval densities remain approximately unchanged (dropping from about 8.5 to about 8 ppg) as indicated at 2106. Moreover the ETOFL of the lowermost interval the first spiked to a positive value before dropping to about −10,000 feet as indicated by the wraparound at 2108. These results strongly indicate a lost circulation event in the lowermost interval, likely at the bit. Drilling and circulation was subsequently suspended.
Managed Pressure Drilling Choke Adjustments
During managed pressure drilling (MPD) operations, the surface annular back pressure (SBP) is maintained such that the bottom hole pressure (BHP) remains in a pre-defined small range in order to prevent both lost circulation and kicks or wellbore stability issues. For example, as the mud pumps are brought down, the surface annular back pressure may be increased in order to compensate for the loss of annular friction and is also adjusted (up or down) to account for possible phase changes when using aerated (or nitrogenated) drilling fluid. Automated feedback control is desirable in order to make the adjustment more timely and accurate. Moreover, automatic control may be further desirable in the event of drilling condition changes (e.g., a kick or change in cuttings density). The back pressure calculations disclosed herein may provide for such automated feedback.
In this operation the goal was to minimize the pressure overshoot and reduce the pressure to the drilling value. The overshoot was reduced by lowering the back pressure over the following 10 minutes (from 23:10 to 23:20) as indicated at 2408. In this operation, a back pressure of about 525-550 psi appears optimal for compensating for the loss of annular friction losses. Therefore, the annular pressure losses due to friction were about 175 psi, rather than the 275 psi originally assumed. Such calibration of the back pressure may improve stability and eliminate inflow issues at connections.
Track 8 displays the computed BP. These computed back pressures indicate the efficiency at which the SBP is being transmitted to the drilling fluid in the annulus at any particular interval. The computed BP may be compared directly in a control loop to obtain a desirable SBP, for example, via adjusting the SBP such that the SBP and computed BP are approximately equal. Since a constant BHP is desirable, the MA_BP—003—001 data may be used directly in the control loop. In
The above described methodology for controlling back pressure during managed pressure drilling operations may be advantageously highly stable since the computed back pressure (from Equation 21) is sensitive to the transmission efficiency of the applied SBP to the annular fluid.
In maintaining a desired BHP during MPD operations, the input flow rate may be adjusted, the mud weight may be adjusted, the volume of injected nitrogen varied, or the BP may be adjusted. In many cases two or more of these parameters may be adjusted substantially simultaneously. Moreover, the average calculated annular BP or any one of the interval calculated BP or the measured downhole measured annulus pressure may be used in an automatic choke control methodology. The choke position may be controlled, for example, in incremental steps by an electro-mechanical device until the system stabilizes and BP and SBP are substantially equal as described above.
Table 10 lists the direction of change for the theoretical BP calculation across the depth intervals while certain other drilling events take place (other than compensating for annular friction losses as described above). These events are listed in column 1. Column 2 lists the desired change in the surface BP during MPD operations in order to counter-act the event down-hole and to maintain a substantially constant BHP (or to maintain the BHP within a safe mud weight window).
As described above, the internal ASM pressures and temperatures may be used to measure the input mud density and temperature profiles. The internal ASM measurements may be further used to compute hydraulic modeling parameters that are in turn used to predict subsequent pressure and temperature effects on the annular fluid as it moves up the annulus. When changing the mud weight or other properties such as the viscosity during a viscous sweep, it may be beneficial to know where the viscous mud (or pill) is in the system. When the mud becomes uniform within the system, drilling can resume.
A circulating time or bottoms up time may be used to determine the depth from which the cuttings collected at the surface have come. Many times the driller will circulate “bottoms up” before POOH (Pull Out Of Hole). This is estimated using an estimated borehole diameter and volume which can be in error. Since the time needed to clean the borehole of all cuttings is not well defined, a safety factor of 1.5 to 2 is commonly used, meaning that circulation time is increased by these factors to insure a clean hole before POOH.
The interval densities and annular friction tend not to change with time once the mud is homogeneous. Non-changing interval densities may therefore be used to determine when the mud density is homogeneous within the borehole volumes. When the annulus is free of cuttings, the annular interval densities tend to reflect the density of the input mud corrected for pressure and temperature effects. Circulation can then be stopped in order to POOH. Either or both of Equations 22 and 23 may be used to determine when the mud system is homogeneous and other drilling operations have resumed.Production Analysis
Obtaining production in wells, especially lateral wells, is often complicated by conveyance issues. In a lateral well, deployment of downhole tools through standard gravity descent may not be possible. To overcome this difficulty, the tools may be either pushed or pulled into the well by means of drill pipe assisted logging, tubing conveyance, tractored, propelled with a swab cup, or some other means. The accumulation of debris while conveying various production tools into the well can be particularly problematic in horizontal or near horizontal wells. Moreover excessive rig time is often required for conveying conventional wireline (WL) tools into horizontal wells such that WL tools are sometimes not used.
Wireline conveyed production analysis tools often include numerous measurement sensors deployed at various depths in the wellbore. Such measurement sensors may alternatively be deployed using wired drill pipe conveyance. The use of WDP enables substantially identical sensors to be deployed in the same configuration and at multiple depths in the wellbore. Sensor deployment may be accomplished via tripping the WDP into the bore hole. The surface pressure may be adjusted such that formation fluids flow into the wellbore and up the interior of the drill pip where they may be vented through a surface choke or routed to production facilities. The along string pressure and temperature measurements as well as the computed interval densities and temperature gradients may then be used to gauge the type and rate of fluid flow from the various intervals. Additionally, by controlling the up-hole pressure, the effect of the pressure variability on the fluid properties down-hole can be assessed—such as phase changes, flow rate changes, liquid holdup changes, and the like.Cuttings Transport Management
Adequate transport of cuttings from the drill bit to the surface is necessary in order to prevent various drilling problems such as friction caused by the accumulation of the cuttings, generation of a pack-off around the BHA or other locations on the drill string, and stuck drill pipe. Increased friction due increased cuttings volume or barite sag in the drilling fluid can slow the removal of the cuttings and result in one or more of the above problems. Cuttings transport issues, if not properly identified and mitigated, can quickly spiral out of control, for example, from increased friction, to a pack-off, to a stuck drill pipe.
In high angle wells, for example including horizontal and near horizontal wells, there is an increased tendency for cuttings to drop out of suspension. This can occur for at least two reasons, including the non-uniform annular flow profile with stagnation increasing towards the bottom of the borehole and the action of gravity in a perpendicular direction to the flow velocity. Having only a short distance to fall into the stagnation flow profile at the bottom of the bore hole, the aforementioned cutting transport problems can therefore manifest quickly in high angle wells.
Various factors such as drill string rotation rate, drilling fluid flow rate, and periodic BHA and drill-pipe axial movements help to keep the cuttings bed stirred up and in suspension. However, at the time of this disclosure there is no known definitive down-hole measurement available to measure the degree of success of these practices at specific depth intervals. Drilling personnel often wait to determine whether or not targeted cuttings appear at the shale shakers approximately (e.g., 20-90 minutes after penetration of the particular formation). Current practice may also make use of single sensor BHA measurements from which drilling personnel look for increases in overall annular density with time to detect cuttings buildups. However, such a buildup may also be due to drilling denser rock with a high rate of penetration or to pack-offs located above the BHA. It is commonly assumed that a decrease in annular density with time corresponds to better hole cleaning and cuttings transport. In reality, cuttings dropping out of solution can give the same signature. In contrast, the ASM pressure and temperature measurements, computed interval densities, and their derivatives may be used to distinguish cuttings drop-out from other effects and locate the affected depth intervals.
An automated routine may be utilized to identify and quantify the severity of a cuttings transport issue (e.g., dropped cuttings from the annular volume) as a function of time and depth prior to running the aforementioned minimization routine. When cuttings are dropping out of suspension, MA_ISD decreases below EA_ISD and approaches (or is substantially equal to) EAF_ISD (as can be seen by comparing
While the interval density changes tend to mimic those of a kick signature and/or a lost circulation signature, cuttings transport issues can be readily identified by noting that Qx=0 in
It will be understood that cuttings transport issues, especially in inclined wells, may cause corresponding changes in certain of the disclosed parameters other than those described above with respect to
A driller may elect to respond to cuttings transport issues, such as cuttings falling out of suspension in the annulus, using a number of mitigating techniques. For example, a drilling operator may elect to (i) increase the rotation rate of the drill string to promote turbulent mixing of the annular fluid, (ii) increase the drilling fluid flow rate, (iii) reduce the rate of penetration (e.g., via reducing weight on bit), or even (iv) replace the drill bit with a less aggressive bit or a bit having a different nozzle configuration. Other BHA components may also be replaced so as to change the pressure drop between the surface and the drill bit. The disclosed embodiments are not limited in any of these regards.Internal and External Temperature Gradients
Internal and annular temperature measurements made as a function of depth and time may be used to compute various temperature gradients in the borehole. For example, internal and external (annular) temperature gradients may be determined along the length of the drill string (as a function of measured depth). Moreover, radial gradients through the drill string between internal and external temperature measurements may be determined These temperature gradients may be utilized to evaluate various drill string and tool related conditions as well as various formation related conditions.
In one embodiment, temperature gradients may be computed as a function of both time and depth along the drill string to predict when the borehole temperature in the BHA may exceed rated tool temperatures. These measurements may be made in both circulating and static conditions. In a high temperature formation the temperature of the borehole may increase with both time and depth during static conditions. Therefore, measured temperature gradients may enable the determination of a time at which rated tool temperatures are exceeded. For example, LWD formation fluid sampling operations are generally carried out during static conditions. The aforementioned temperature gradients may enable a maximum time-on-station to be determined during which the sampling operation would need to be completed. Circulation may then be resumed so as to cool the BHA.
In another embodiment internal and external measurements may be used to model a radial heat transfer coefficient of the drill string or downhole tool. Such modelling may further include a third temperature measurement to be made between the internal and external fluids (e.g., in an internal circuit board). The use of three temperature measurements may enable non-linear heat transfer effects to be evaluated. Such measurements may be made during circulating and/or static conditions. These temperature measurements may be included in a model to predict drill string temperatures for numerous drilling conditions. For example, temperature gradients may be evaluated at multiple drill string rotation rates (e.g., 50 rpm, 100 rpm, and 200 rpm) and at multiple drilling fluid flow rates (e.g., 300 gpm, 500 gpm, and 800 gpm). This may enable the effects of various drilling parameters, including drill string rotation rate and drilling fluid flow rate, in mitigating high temperature drilling situations.
Developing a heat transfer model, for example, as described in the preceding paragraph may further enable the measured temperatures to be used to calculate a static formation temperature. Obtaining the static formation temperature may be highly valuable in that it is related to numerous parameters of interest including formation heat transfer capacity which is in turn related to the fluid and lithology content of the formation which is still further related to the porosity, hydrocarbon saturation, and pore pressure. Determination of the static formation temperature may further enable circulating and static borehole temperatures to be predicted long before completing the well. Phase changes may also be identified. Moreover knowledge of the static formation temperature may enable staging plans to be refined while tripping into hot wells.
Although numerous methods for computing and utilizing wellbore interval densities and certain advantages thereof have been described in detail, it should be understood that various changes, substitutions and alternations can be made herein without departing from the spirit and scope of the disclosure as defined by the appended claims.
1. A method for estimating an interval density in a subterranean wellbore, the method comprising:
- (a) deploying a tool string in the wellbore, the tool string including at least first and second subsurface longitudinally spaced pressure sensors deployed at corresponding first and second measured depths in the wellbore;
- (b) causing the first and second pressure sensors to acquire first and second drilling fluid pressure measurements at the first and second measured depths; and
- (c) causing a processor to process the first and second pressure measurements to compute an interval density between the first and second measured depths in the wellbore.
2. The method according to claim 1, wherein: the first and second pressure measurements are transmitted to the surface and the interval density is computed at the surface.
3. The method according to claim 1, wherein:
- the tool string includes at least first, second, and third subsurface longitudinally spaced pressure sensors deployed at corresponding first, second, and third measured depths;
- (b) comprises causing the at least first, second, and third pressure sensors to acquire at least first, second, and third drilling fluid pressure measurements; and
- (c) comprises processing the at least first, second, and third pressure measurements to compute at least first and second interval densities, the first interval density between the first and second measured depths and the second interval density between the second and third measured depths.
4. The method according to claim 1, wherein:
- the tool string includes at least n=3 longitudinally spaced pressure sensors;
- (b) comprises causing each of the n pressure sensors to acquire a drilling fluid pressure measurement; and
- (c) comprises processing the n pressure measurements to compute N interval densities, wherein n−1≦N≦Σi=1i=n−1(n−i).
5. The method according to claim 1, wherein the pressure sensors comprise annular pressure sensors.
6. The method according to claim 1, wherein the pressure sensors comprise internal pressure sensors.
7. The method according to claim 1, wherein the tool string comprises wired drill pipe.
8. The method according to claim 1, wherein drilling fluid is being circulated in the wellbore in (b).
9. The method according to claim 8, wherein a drill bit is rotating on bottom in (b).
10. The method according to claim 1, wherein drilling fluid is substantially static in the wellbore in (b).
11. A method for drilling a subterranean wellbore, the method comprising:
- (a) deploying a drill string in the wellbore, the drill string including at least first and second subsurface longitudinally spaced annular pressure sensors deployed at corresponding first and second measured depths in the wellbore;
- (b) causing the first and second annular pressure sensors to acquire first and second annular pressure measurements at the first and second measured depths;
- (c) processing the first and second annular pressure measurements to compute an annular interval density between the first and second measured depths; and
- (d) evaluating the annular interval density computed in (c) as an indicator of an adverse drilling condition.
12. The method according to claim 11, wherein the adverse drilling condition is selected from the group consisting of inflow from a subterranean formation to the wellbore, outflow from the wellbore to a subterranean formation, borehole washout, borehole pack-off, poor cuttings transport, and tar intrusion into the wellbore.
13. The method according to claim 11, wherein the drill string comprises wired drill pipe.
14. The method according to claim 11, wherein:
- (b) further comprises causing first and second internal pressure sensors to acquire first and second internal pressure measurements at the first and second measured depths;
- (c) further comprises processing the first and second internal pressure measurements to acquire an internal interval density between the first and second measured depths; and
- (d) further comprises evaluating the internal and annular interval densities computed in (c) as indicators of an adverse drilling condition.
15. The method according to claim 14, wherein:
- drilling fluid is being circulated in the wellbore in (b); and
- (c) further comprises computing (i) a static annular interval density via subtracting a modeled annular friction component from the annular interval density and (ii) a static internal interval density via subtracting a modeled internal friction component from the internal interval density.
16. A method for drilling a subterranean wellbore, the method comprising:
- (a) deploying a drill string in the wellbore, the drill string including at least first and second subsurface longitudinally spaced annular pressure sensors deployed at corresponding first and second measured depths in the wellbore;
- (b) circulating drilling fluid through the drill string;
- (c) measuring an annular interval density between the first and second pressure sensors;
- (d) comparing the annular interval density measured in (c) with a modeled annular interval density to obtain a difference; and
- (e) evaluating the difference obtained in (d) as an indicator of an adverse drilling condition.
17. The method according to claim 16, wherein the adverse drilling condition is selected from the group consisting of inflow from a formation to the wellbore, outflow from the wellbore to a formation, borehole washout, borehole pack-off, poor cuttings transport, and tar intrusion into the wellbore.
18. The method according to claim 16, wherein the modeled annular interval density is obtained via:
- (i) acquiring an internal interval static density measurement;
- (ii) processing the internal interval static density measurement to adjust hydraulic model parameters; and
- (iii) using the hydraulic model parameters to compute the modeled annular interval density.
19. The method according to claim 18, wherein:
- (i) further comprises acquiring an annular interval static density measurement;
- (ii) further comprises processing the annular interval static density measurement to adjust the hydraulic model parameters; and
- (iii) using the hydraulic model parameters to compute the modeled annular interval density.
20. The method according to claim 16, wherein
- (c) further comprises computing a static annular interval density via subtracting a modeled annular friction component from the measured annulus interval density;
- (d) further comprises comparing the static annular interval density computed in (c) with a modeled static annular interval density to obtain a interval static density difference; and
- (e) further comprises evaluating the interval static density difference obtained in (d) as an indicator of an adverse drilling condition.
21. The method according to claim 16, wherein:
- the drill string includes at least first, second, and third subsurface longitudinally spaced annular pressure sensors deployed at corresponding first, second, and third measured depths;
- (c) comprises measuring a first annular interval density between the first and second measured depths and a second annular interval density between the second and third measured depths;
- (d) comprises comparing the first and second interval densities measured in (c) with first and second modeled annular interval densities to obtain first and second differences; and
- (e) comprises evaluating the first and second differences obtained in (d) as indicators of an adverse drilling condition.
Filed: Aug 14, 2012
Publication Date: Feb 28, 2013
Inventors: John Rasmus (Richmond, TX), William Lesso (Anderson, TX), John James (Houston, TX), Edward M. Tollefsen (Katy, TX), Scott Paul (Houston, TX), Amanda L. Weber (Oklahoma City, OK), Marcus Turner (Sugar Land, TX), Paul Bolchover (Beijing)
Application Number: 13/585,495
International Classification: E21B 47/00 (20120101); E21B 47/06 (20120101);