• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    Tools, systems, and methods may be configured for or include one or more of a combination of acoustic and acoustic data. This paper is based on the analysis of spatial and temporal data.
    公开号:FR3037993A1
    申请号:FR1654723
    申请日:2016-05-26
    公开日:2016-12-30
    发明作者:Yi Yang Ang；Nam Nguyen；Joni Lie
    申请人:Halliburton Energy Services Inc；
    IPC主号:
    专利说明:

    [0001] BACKGROUND OF THE INVENTION [0001] The present application relates to the detection of leakage in a wellbore. BACKGROUND OF THE INVENTION [0002] During the life cycle of a gas and oil well, it is desirable to monitor and maintain the integrity of the well. In particular, the integrity of the well barriers (such as production lines, well casing and surrounding cement sheath) is important to ensure a safe operation of the well and to avoid burst or leakage events. hydrocarbons in the environment. Leaks at the well barriers can, in principle, be detected by monitoring underground fluid flows (eg, gas or oil) in and around a wellbore. Monitoring of downhole flows around boreholes, such as injected water, may also be important during tank characterization. BRIEF DESCRIPTION OF THE DRAWINGS [0003] The following figures are presented to illustrate certain aspects of the embodiments, and should not be construed as exclusive embodiments. The subject matter of the disclosed invention can undergo considerable modifications, alterations, combinations and equivalents in form and function, as will be apparent to those skilled in the art who benefit from this disclosure. [0004] FIG. 1 illustrates an example of an acoustic sensor array deployed inside the wellbore. [0005] FIG. 2 illustrates, at a high conceptual level, the way in which an acoustic source can be located in two dimensions. [0006] FIG. 3 illustrates an overall flow diagram of a moving beam forming method that uses modeled steerable vector weights depending on an adaptive time. FIG. 4 is an illustrative graph showing an example AT for the various network speeds. Figure 5 illustrates an example of a cable logging system in a well. The system includes a probe tool suspended from a cable in a cased wellbore. FIG. 6 illustrates a system for the detection and localization of underground acoustic sources implemented in a drilling system for MWD logging operations. [0010] FIGS. 7A-7C compare the accuracy (A) of a stationary sensor array with independent steered vector weightings of a fixed duration, (B) a removable sensor array with weightings of modeled steerable vectors independent of a fixed duration and (C) of a detachable sensor array with modeled steered vector weightings dependent on an adaptive duration in which the source frequency (7 kHz) is lower than the frequency of the spatial folding. FIGS. 8A-8C compare the accuracy (A) of a stationary sensor array with modeled steerable vector weightings independent of a fixed duration, (B) of a removable sensor array with vector weights. orientable steerable models independent of a fixed duration and (C) of a removable sensor array with modeled steered vector weightings dependent on an adaptive duration in which the source frequency (25 kHz) is greater than the fold frequency spatial. FIGS. 9A-9C compare the accuracy (A) of a stationary sensor array with independent modeled steerable vectors weightings of a fixed duration, (B) of a removable sensor array with weightings of modeled steerable vectors independent of a fixed duration and (C) of a removable sensor array with modeled steered vector weightings dependent on an adaptive duration in which the source frequency (47 kHz) is lower than the fold frequency spatial.
    [0002] DETAILED DESCRIPTION [0013] The present application relates to leak detection in a wellbore. Subterranean fluid streams generally emit acoustic signals that can be measured, for example, with fiber cables deposited along the wellbore or with acoustic point sensors such as hydrophone sensors or hydrophones. a network of Bragg fibers (FBG). Existing methods are, however, very limited in their accuracy and the accuracy with which a detected fluid flow can be located. In addition, existing methods generally assume or require that the tool be relatively stationary when recording acoustic signals from the fluid streams. However, in practice, the tool preferably travels through the wellbore generally at a fixed speed of about 20 feet per minute to about 30 feet per minute. The exemplary embodiments described herein include tools, systems, and methods for detecting one or more underground acoustic sources and locating them according to the depth distance and the radial distance from a wellbore. utilizing a network of at least 3 acoustic sensors (also referred to herein as a "sensor array") placed in the wellbore in conjunction with a network signal processing which accounts for the movement of the acoustic sensors and the folding effects using model-dependent time-dependent adaptive vector weightings. The detection and location of the acoustic source in accordance with the present invention can be used, in particular, to identify underground fluid streams (eg, from leakage in the well barriers) from which the acoustic signals emanate. [0016] As used herein, the term "depth" of the present invention generally describes a coordinate along a direction of a wellbore, regardless of whether the borehole extends vertically into a wellbore. the formation or that it is inclined with respect to the vertical direction. As used herein, the term "radial distance" describes a direction perpendicular to and away from the longitudinal axis of the wellbore. As used herein, the term "network signal processing" generally describes techniques for estimating or calculating the parameters of one or more signal sources (such as source locations). and emitted waveforms) by merging the data collected by a sensor array having known geometric relationships, either substantially or simultaneously or, more generally, having known time relationships between the different sensor signals. As used herein, the term "substantially simultaneously" with respect to time intervals indicates that the time intervals during which the signals are collected substantially overlap (e.g., by at least 90%) by preferably at least 99%) between the different sensors. The network signal processing techniques include, without limitation, various spatial filtering methods, such as conventional beamforming, Capon beamforming, multiple signal classification (MUSIC), and various parametric methods. as well as the estimation of the delay time. Network signal processing usually depends on an advanced pattern of wave propagation from there or sources to the sensors to solve the inverse problem (eg, source location). In conventional application contexts, such as radar and sonar, this advanced model is generally simple since the propagation of the wave is in a uniform (homogeneous or isotropic) medium (e.g. air or in water) and we can assume that the source is far from the sensors. However, when fluid flows in or around a wellbore are measured, the uniform medium and distance assumptions no longer hold. Therefore, in various embodiments, the advanced model is adjusted to take into account the configuration and condition of the wellbore and the surrounding formation (which collectively include various propagation media and boundaries between the two) and their effect on the wave field (eg, wave refraction, reflection and resonance), as well as to facilitate the processing of near-field signals (ie, signals from a source whose distance from the sensors is not significantly (e.g., order of magnitude) larger than the spatial extent of the sensor array). The implementation of the network signal processing involves, in accordance with some embodiments, merging the signals received by the individual network sensors for the plurality of putative source locations within a two-way region. predefined size (which, for example, covers a certain length in the direction of the depth and extends a certain radial distance from the borehole) in order to calculate a 2D map of an energy level, an amplitude of the acoustic source, or other merged signal parameters as a function of depth and radial distance. The actual locations of the source can be determined from this map by identifying the local maximum (or local multiple maxima) of the energy level of the acoustic source or other parameter. The magnitude of the local maximum can be used to infer whether or not the identified acoustic source really corresponds to an underground flow. For example, in some embodiments, the acoustic signals are acquired under multiple flow and non-flow conditions to establish a statistical detection threshold for the streams for use in the binary hypothesis test or a similar statistical test. [0020] FIG. 1 illustrates an example of a removable acoustic sensor network 10 deployed inside a wellbore, in accordance with the various embodiments. As demonstrated, the sensors 100 can be placed linearly along the longitudinal axis 102 of the wellbore (whose radial coordinate is zero). They may be evenly spaced (as illustrated) or there may be varied spacing between adjacent sensors. The sensor environment generally comprises multiple physical barriers to fluid flow, such as production pipes 104 through which gas or oil may be pumped up and out of the well, one or possibly multiple casings of fluid. nested wells 106 and a cement sheath 108 filling the space between the casing (s) 106 and the formation 110 surrounding the wellbore. In addition, the wellbore can be divided into multiple vertical sections, eg by forming joints 112 between the casings 106 which can separate, for example, a perforated lower portion of the casing through which the hydrocarbons enter. from an upper part (not perforated) serving as a conduit upwards. Unexpected flow scenarios that may occur in this configuration may include, without limitation, flows through casing 106 or pipes 104 due to cracks or holes therein (identified by arrows 120). the flows through a forming joint 112 between adjacent vertical sections of the wellbore due to insufficient sealing (identified by the arrows 122) and flows within the formation 110, the cement sheath 108 or another layer that is more or less parallel to the boundaries of the layer (identified by arrows 124). When flows flow through restricted paths, acoustic signals can be generated due to the pressure drops that accompany these flows. The acoustic signals propagate generally in all directions through the formation and / or borehole, which can be detected at the various sensor locations. Acoustic sensors 100 suitable for use in the embodiments described herein include, e.g. and without limitation, hydrophones (piezoelectric), FBG sensors or segments of a distributed optical fiber cable. In various embodiments, the acoustic sensors are omnidirectional, i.e., unable to differentiate by themselves the different incoming directions of the signal. By exploiting the spatio-temporal relationships between signals from the same source at multiple sensors, however, information regarding direction and / or location of the signal source can be obtained. For example, using at least three sensors in a linear arrangement along the axis of the wellbore, as shown in FIG. 1, it is possible, at least under certain conditions, to determine the depth and the radial distance of the source (as will be described in more detail below). In order to also locate the source in the azimuthal direction, the configuration of the sensor array can be varied, e.g. by placing different sensors at different radial positions by placing them in two or three dimensions, partially overlapping the sensors to to limit their detection to certain azimuthal windows (different windows for different sensors) while pivoting the partially covered sensors to cover the entire azimuthal region, or using steerable sensors (i.e. sensors that inherently provide directional information). A linear configuration illustrated in FIG. 1 may be a consequence of the spatial confinements imposed by the pipes 104 in which the sensors are mounted. FIG. 2 illustrates, at a high conceptual level 200, the manner in which an acoustic source 202 (e.g., a fluid flow) can be located in two dimensions (e.g., a radial distance X and a depth y) based on the signals coming simultaneously from multiple sensors 204, 206, 208 (shown as three sensors) at different locations R1, R2 and R3, respectively, provided that the medium is uniform so that the signal travels from the source to the sensors along straight lines (without undergoing, eg, refraction or reflection) and at a known and constant speed, the sound y. In this case, the travel time t of the signal from the source to a sensor is simply the ratio of the distance d 210,212,214 between the source 202 and the sensors 204,206,208, respectively, to the V. As it will be easily understood by one skilled in the art, the absolute t can not be measured by passive flux detection methods described herein since the acoustic signal does not have a known starting point in time (since the flux usually starts long before measurements and, in any case, at an unknown time). However, the time delay Atij = ti - tj between the reception of a certain characteristic of the signal (eg a peak in the time waveform) at a first sensor i and the reception of the same The characteristic at a second sensor j (i.e., the relative phase shift) can, in principle, be determined with equations 1 and 2 d., (Xs -Xri) 2 ( Ys-Yri) 2 ÉQ. 1 t- = = 15 atii = ti - = xs-xri) 2+ (Ys-Yri) 2- 1 (xs-xrj) 2 (.Ys-Yrj) 2 EQ 2 With known locations of sensors (R1 to (xrifyri), R2 to (Xr2 yr2) and R3 to (Xr3 yr3) and a known, this delay time gives a non-linear equation containing two unknowns, including the coordinates (xs tys) of the source s A second time delay measured between one of the sensors i or j and a third sensor k gives a second independent nonlinear equation, and from these two equations, the location of the source in 2D can be calculated simply. In a manner known to those skilled in the art, if the v is unknown and / or changes as the signal propagates through different media, a network with a large number of sensors (e.g., 4 or more sensors) may be used to provide enough information to locate the source 3037993 8 [0024] In the most complex scenarios generally found in d applications detection of flows, as envisaged here, the signal processing generally takes a more complex form. In various embodiments, a signal processing method of a network (such as spatial filtering) may be used to merge the various simultaneously acquired sensor signals and locate the acoustic source. In some cases, the network signal processing technique may comprise at least one of a spatial filtering, an estimation of a delay time or a sound-energy method. FIG. 3 is an overview of an example of a network signal processing technique 300 which uses modeled steered vector weightings dependent on adaptive time 316, according to some embodiments of the present application. For example, a narrow-band ("s (t)") remote acoustic source may be used, as described herein, to illustrate the principles underlying the spatial filtering methods of the present disclosure. Mathematically, the signal 302 picked up using the static sensor array placed along the y-axis (i.e., along the longitudinal axis 102 of the borehole of the 1) can optionally be normalized by applying a normalization operation 304 to produce a received signal Pm (t) 306 at the level of the mth sensor with signals from the leak source K, if (t), ..., sK (t). Since there are M sensors in the array with inter-element spacing d, the received signal vector 306 of each sensor can be expressed as EQ 3. 25 pm (t) = E f = 1 ain (rsk) sk (t) + chn (t), EQ 3. in which am (rsk) is the transfer function of the k th source propagating towards the mth sensor, rsk = (Xsk, ysk) is a vector identifying the position of the kth incident source, Sk (t) represents the magnitude of the kth incident source and qm (t) represents the additive Gaussian white noise that captures the effects of thermal and environmental noise. The pm (t) signal 306 from all the sensors may also be represented as a vector p (t) as defined in the QoE. 4. Kt) = [1 (t) Pm (t) Pm (t) 11. = As (t) + q (t), Éq 4. 10 where s (t) = [s1 (t) sm (t) siat) iT A = [a (r si) ... a (rsk) i , q (t) = [q1 (t) ... ch (t)] T, a (rsk) = [ai (rsk) am (rsk) ... am (rs')] T, and (.) T identifies the transposition operation. With the EQ 4 the theoretical covariance matrix R of the static sensor array model has the following form. R = AE ts (t) sl (t)} All + aqI, where E {. } identifies the wait operation, (.) II identifies the Hermitian transposition operation, and Oc / represents the assumed power of the noise I being the identity matrix. In practical applications, the theoretical covariance matrix R can be replaced by a sample of covariance matrix calculated using the QE. 6. 1 t = 1T EQ 6. R = f p (t) pH (t) dt. Where AT represents the duration of the received signal received 306 for the network. Conventionally, the location of the static sensor array can be achieved by constructing the spatial spectrum p (r;) and looking for the peaks. Peaks are used as the probability of source location. By letting 15 = {T1 TL} identify the set that contains the grid location in which the spatial scan is performed, then, for each location, the independent fixed time modeled steered vectors weights a (ri) where r - E R. is first calculated, before estimating the standard Spatial Spectrum Capon pfri) by the QE. 7. 13 (r 1) = a 1 (77) 111a (r 1). EQ. Ideally, if the network is stationary, the local maximum will occur ri = rsk when the weight of the orientable vector a (ri) matches well with the covariance matrix R. [0033] However, when the network As the number of sensors moves at a constant velocity V, the received signal 306 in the duration of AT will produce a covariance matrix t which no longer corresponds to the weight of the orientable vector a (ri) of the stationary model in the QE. 5. Specifically, the relative displacement caused by the motion at time t (0 <t <An can be modeled as Ar = vLT, in which case the received signal 306 from the K sources can be expressed as: p (t, Ar) = Ek 1 a (rsk ± Ar) sk (o + q (t) = A (Ar) s (t) + ci (t) where E (3, r) [a (3) rsi + Ar) ... a (rsk + Ar)] - [0034] Taking into account the moving sensor array, the present application aims to locate the position of the source of the leak rsk = 15 (xsk, ysk) where k = t1,, K} by following the analysis of the received signal 306. When the theoretical covariance matrix from the received signal vector in the EQ 8 is identified as R (Lr), EQ 9 is derived R (Ar) = Efp (t, Ar) p (t, Ar)) = A (Ar) Efs (t) sH (t)} All (Ar) + QqI. Eq. 9 20 When t, V or the frequency of the signal increases (shorter wavelength), the difference between A (Ar) and A also increases, and therefore the direct application of the EQ 7 using the covariance matrix generated in time tT with the QE. 9 will cause a mismatch at its orientable vectors.
    [0003] An ad hoc approach to solving this mismatch is to shorten the AT acquisition duration of the QE. 6 until the stationary condition is preserved. In practice, this can be achieved by truncating AT and using a time-weighted, fixed-weighted modeled steerable vector, which is a constructional operation of the Capon Spectrum 318, described later. However, in order to effectively resolve the mismatch, the solution is given by dividing the received signal 306 recorded within the acquisition window AT into multiple frames. In general, when a frame decomposition operation 308 is applied to the received signal 306, the received signal 306 decomposes into shorter equal decomposed N frames 3100-310N-1 in which the condition is stationary and preserved. Then, each of the decomposed frames 3100 ... 310N-1 represents a virtual network. For the set of N decayed frames 3100 ... 310N-1, the corresponding theoretical covariance matrices R (ntr) can be represented by the QE. 10. tR (O) R (Ar) -R (nAr) - - R ((N-1) Ar)). EQ. The application of a covariance matrix sample estimator 3120 ... 312N-1 (EQ.11) to each of the decomposed frames 3100 ... 310N-1 converts each of the 3100 decomposed frames. .310N-1 in a sample of covariance matrix 3140 ... 314N-1 (tI (0) --- (nAr) - - 1 - ((N-1) Ar))). (nAr) = - Jt = ned IN p (t, nzr) pH (t, nAr) .N- ft = (n + 1) AT / N EQ. [0039] By applying the Capon spatial spectrum construction operation 3180 ... 318N_1 (Eq.12), the weightings of the adaptive modeled adaptive vectors dependent on the duration 316o ... 316N-1 at one any single sample of covariance matrix rk (nAr) 3140 ... 314w1, the response of the Capon spatial spectrum (13 (71) (r1)) 320o ... 320N-1 for each of the frames is produced, in which the peak signals corresponding to the source location can be searched for.
    [0004] R (n) (r) = b EQ. 12) where r is the weighting factor for each nth frame and a (r1 + nr) is the weighting of the adaptive modeled adaptive vectors that are dependent on the nth (rj + n1r) ii-1 (Ar). the duration 3160 ... 316N-1. [0040] Even if any response of the individual Capon Spectrum r3 (n) (r1) 3200 ... 320N-1 can be used to search for the source location rsk = (xsk, ysk), the operation of Capon Spectrum Construction 3180 ... 318N_1 may be subject to spatial folding that produces side-lobe peaks in the Capon Spectrum Response 3200 ... 320N-1, which could be erroneously identified, as source locations. In order to reduce the side-lobe intensity, a weighted average response of the Flooving (1 "j) 324 Spectrum Spectrum is calculated by performing a normalization and addition operation 322 (EQ 13) on the spectrum response. Capon spacecraft 320o ... 320N-1 Since the side lobes are not consistently located in each of the Capon Spectrum 3200 ... 320N_1 responses, the normalization and addition operation 322 results in a reduction in the intensity of side lobe peaks and an increase in the intensity of the source location 3037993 14 fi'moving (ri) - p (n) (7.1) EQ 13 [0041] In some cases acoustic signals from acoustic sources may have a higher frequency than a spatial folding frequency used in the network signal processing technique, and acoustic signals may have a frequency lower than a replication frequency. Spatial nt used in the technique of network signal processing. For example, to prevent spatial folding, the inter-sensor spacing may be d_ 212. Where To. represents the wavelength of the incident signal. However, since the network is designed with a fixed inter-sensor space, it can suffer from a spatial folding effect when the wavelength of the incident signal is λ / 2 d. In some cases, the QE. It can be used to link the design of ATn = AT IN, where: when the grating moves only along the z axis at a velocity, V = Vz, then we can approach ATn as stationary if the difference between the radial distance of nLT and (n - 1) AT is related to the inside of 1%. The 1% binding stress can be mathematically expressed as the QE. 14.
    [0005] ATn <EQ. Fig. 4 is an illustrative graph showing an example ATn for different network speeds with a 1 ° h connection. [0044] The methods described herein may be implemented by a set of instructions that causes a processor to perform the network signal processing technique described herein which comprises a time-dependent adaptive modeled steerable vector for determining depth and a radial distance from a wellbore of the acoustic source. The processor may be part of a computer hardware used to implement the various illustrative blocks, modules, elements, components, methods, and algorithms described herein. The processor may be configured to execute one or more instruction, programming position, or code sequences stored on a computer-readable non-transitory medium. The processor may be, for example, a versatile microprocessor, a microcontroller, a digital signal processor, an application specific integrated circuit, a programmable gate array, a programmable logic device, a control, a state machine, a logic gate, discrete hardware components, an artificial neural network, or any such appropriate entity that can perform calculations or other data manipulations. In some embodiments, a computer hardware may include such elements as, eg, a memory (eg, RAM, flash memory, ROM, PROM, EPROM, registers, hard disks, removable disks, CD-ROMs, DVDs, or any other suitable similar storage device or medium [0046] The executable sequences described herein may be implemented with one or more code sequences contained in a memory. In some embodiments, such a code may be read into the memory from another machine-readable medium.The execution of the instruction sequences contained in the memory may cause the processor to perform the process steps described herein. One or more of the processors in a multiprocessor array may be used to execute the instruction sequences in the memory, and a hardwired circuit may be provided. t be used in place of or in association with software instructions to implement various embodiments described herein. Thus, the present embodiments are not limited to a specific combination of hardware and / or software. In this context, a computer readable medium is any medium that directly or indirectly transmits instructions to a processor for execution. A computer readable medium may take any form including, for example, a non-volatile medium, a volatile medium, and a transmission medium. Volatile support may include, for example, a dynamic memory. A transmission medium may include, for example, coaxial cables, wires, optical fiber, and wires that form a bus. Common forms of computer readable media may include, for example, floppy disks, flexible disks, hard disks, magnetic tapes, other than magnetic media, CD-ROMs, DVDs, and other media. such optics, punch cards, paper strips and physical media of this type with holes, RAM, ROM, PROM, EPROM and flash EPROM. The detection and location of underground acoustic sources (and thus, underground fluid streams) in accordance with the present invention can be implemented in both cable logging and borehole measurement (MWD) operations. . Figure 5 illustrates an example of a well logging system. The system includes a probe tool 500 suspended from a cable 502 in a cased wellbore 504. In various embodiments, the tool 500 is used within the production casing 506 through which the hydrocarbons are pumped out of the wellbore 504. The tool 500 comprises a plurality (at least three) acoustic sensors 508 (such as, eg, hydrophones), e.g., placed in a linear array 510 along a longitudinal axis 511 of the tool 500 and, therefore, of the wellbore 504. In addition, the tool may comprise an appropriate control and processing circuit 512 which may, in turn, be in communication (e.g. Through a wired connection or a telemetry system) with a surface data processing system 514. The data processing module which provides the computer function for processing and merging the acoustic signals received by the individual sensors. 508 and the d detection and location of streams based thereon may be implemented by a control circuit 25 and processing 512 or the data processing system 514, or both in combination. For example, in some embodiments, the control and processing circuit 512 preprocesses the signals of the individual sensor (eg, by signal conditioning, noise filtering and / or suppression) and transmits them to the data processing system. surface data 514, where the merged signal map is computed, and the flux-induced acoustic sources are detected and located based on it. Each of the control and processing circuit 512 and the surface data processing system 514 can generally be implemented in hardware, or a combination thereof, such as, for example, a programmed multipurpose or dedicated computer of suitably including, for example, a processor and associated memory (as shown in FIG. 6). In various embodiments, the processed acoustic signals are evaluated in combination with measurements from other sensors (eg, temperature measurements and well surface pressure (to evaluate flow conditions and temperature). Overall Well Integrity Alternative sensor configurations may be used to support the detection of the acoustic source in a wireline logging operation, eg, in some embodiments, a distributed optical fiber cable is used instead of acoustic point sensors The fiber optic cable can be permanently installed in the wellbore, eg, fixed behind the casing or integrated in the cemented ring. segment of the optical fiber cable, can be optically scanned to detect surrounding acoustic signals.In this configuration, different channels to 15 different pr ofondeurs correspond to different acoustic sensors. Using a cable logging tool 500, the acoustic sensor array can search, at any drilling depth, for a predefined 2D space, e.g., the length of the grating opening in the direction of the depth and a few meters in the formation in the radial direction. This search can be repeated as the network moves to another drilling depth. Thus, in one passage of the logging cable, a region covering the entire length of the well can be analyzed to identify acoustic sources induced by the flow. In some embodiments, the acoustic sensor array operates at a fast drilling rate (e.g., a speed of up to 60 feet per minute) to detect flows initially at a coarse spatial resolution. Once one or more flows have been detected at certain depths, regions of these depths may be re-examined at lower scan speeds, or in stationary mode, to locate the stream (s) with finer spatial resolution. . In embodiments in which an acoustic signal is emitted along an extended path (as opposed to transmitting from a source point) the integral flux path can be mapped in a space 2D depth and radial distance. [0052] Referring now to FIG. It can be seen that a system for detecting and locating underground acoustic sources can also be implemented in a drilling system for MWD logging operations. This can be useful, for example, for detecting flows in order to characterize formation and hydrocarbon reservoirs and to orient or adjust the drilling as a function thereof. As illustrated, the drilling system comprises a drilling platform 600 located at the level of the surface of the well 604 and, supported by the drilling platform 600, a drill string 606 for drilling a drill. Drill well 608 through subterranean formations 610. Drilltray 606 includes a drill pipe 612 and, generally located at the lower end of drill pipe 612, a downhole module (BHA) 614. The BHA 614 may include the drill bit 616 and, placed thereon, one or more drill collars 618, 620 which may contain a number of different tools and instruments suitable for taking measurements. during the drilling process. In accordance with the various embodiments, these tools may include an array of acoustic sensors 624 (eg, including three or more sensors placed in a linear fashion) and the associated control or processing circuit 626, and may be in communication with one another. Surface Data Processing System 628. Together, the acoustic sensor array 624 and the control and processing circuit 626 and / or the data processing system 628 provide a function for implementing the methods described above. The present disclosure includes Embodiments A-C. Embodiment A is a method which comprises moving a sensor array comprising at least three sensors along a wellbore; measuring, substantially simultaneously, acoustic signals from an acoustic source with each of the at least three sensors; and processing the acoustic signals in association with a network signal processing technique that uses time-dependent adaptive modeled steerable vector weights to identify a location of the acoustic signal by a depth and a radial distance from the wellbore. Embodiment B is a system that includes a sensor array that can be moved along a wellbore and including at least three acoustic sensors for substantially simultaneous measurement of acoustic signals from an acoustic source and which are received at the same; a computer-readable non-transitory medium encoded with instructions which, when executed, perform the operations of: measuring, substantially simultaneously, acoustic signals from an acoustic source with each of the at least three sensors in a network of sensors moving along a wellbore; and acoustic signal processing in conjunction with the use of a network signal processing technique which comprises a time-dependent, modeled adaptive steerable vector for determining a depth and radial distance from the borehole of the present invention. the acoustic source. Embodiment C is a non-transitory computer readable medium encoded by instructions which, when executed, carry out operations comprising: measuring, substantially simultaneously, acoustic signals from an acoustic source with each of the at least three sensors in a sensor array moving along a wellbore; and acoustic signal processing in conjunction with the use of a network signal processing technique which includes a time-dependent, modeled adaptive steerable vector for determining a depth and a radial distance from the borehole of the acoustic source. [0058] Optionally, Embodiment A, B or C may also include one or more of the following: Element 1: wherein the acoustic sensors form a linear array positioned along a longitudinal axis of the wellbore. Element 2: wherein the network signal processing technique comprises at least one of a spatial filter, an estimate of a delay time, and a sound energy method; Element 3: wherein the acoustic sensors comprise one of: an omnidirectional hydrophone, a Bragg fiber grating sensor or an optical fiber cable, and any combination thereof; Element 4: the method or operations also comprising calculating an azimuthal position of the acoustic source; Element 5 wherein the acoustic signals have a frequency higher than a spatial folding frequency used in the network signal processing technique; Element 6: wherein the acoustic signals have a frequency lower than a spatial folding frequency used in the network signal processing technique; Element 7: in which a stationary condition of the sensor array is preserved by satisfying, AT, or AT, is a duration of the acoustic signal, ri is a location of the first sensor of the at least three sensors, and is a speed of the first sensor moving along the wellbore; Element 8: Element 7 and wherein the network signal processing technique comprises: applying a data truncation operation to the acoustic signals to produce frames; applying an estimator of a covariance matrix sample to the frames to produce a covariance matrix sample; applying a Capon spatial spectral construction operation using the weighted, time-independent, model-orientated steerable vector weights to the covariance matrix sample to produce a Capon Spatial Spectrum Response; and identifying the location of the acoustic signal based on the response of the Capon Spectrum; and Element 9: wherein the network signal processing technique comprises: applying a frame decomposition operation to the acoustic signals to produce decomposed frames; applying a sample estimator of the covariance matrix to the decomposed frames to produce a covariance matrix sample; applying the Capon Spatial Spatial Construct operation using model dependent adaptive steerable vectors dependent on the sample duration of the covariance matrix to produce a Capon spatial spectrum response; applying a normalization and addition operation to the Capon Spatial Spectrum Response to produce a Capon Spatial Spectrum Weighted Mean Response and identifying the acoustic signal location based on the average response weighted Capon spatial spectrum. Examples of combinations include, without limitation: two or more of the elements 1-4 in combination; one or more of Elements 1-4 in combination with Element 5 or Element 6; one or more of Elements 1-4 in combination with Element 7 (and possibly Element 8) or Element 9; Element 5 or Element 6 in combination with Element 7 (and possibly Element 8) or Element 9; and combinations thereof. [0059] Unless otherwise indicated, all numbers expressing amounts of ingredients, properties such as molecular weight, reaction conditions, etc. used in this application and the appended claims must be considered in any case to be modifiable by the term "approximately 3037993". Therefore, unless otherwise indicated, the numerical parameters described in the following application and the appended claims are approximations that may vary depending on the desired properties that are desired by the embodiments of the present disclosure. At the very least, and not as an attempt to limit the application of the doctrine of equivalents to the scope of the claims, each numerical parameter must at least be interpreted in the light of significant decimal points and by applying the ordinary rules of rounding . [0060] One or more illustrative embodiments incorporating the embodiments disclosed herein are shown here. Some features of the physical implementation are not described in this application for the sake of clarity. It should be understood that in developing a physical embodiment embodying the embodiments of the present disclosure, many implementation-specific decisions must be made in order to achieve the developer's objectives, such as compliance with system, business, government, and other constraints that vary from implementation to implementation and from time to time. While a developer's efforts may be time-consuming, such efforts would, nevertheless, be a routine task for a skilled person who benefits from the present disclosure. Although the compositions and methods are described herein in terms of "comprising" various components or steps, the compositions and methods may also "consist essentially of" or "consist of" various apparatus or steps. [0062] To facilitate a better understanding of the embodiments of the present disclosure, the following examples of preferred or representative embodiments are given. In no case may the following examples be construed as limiting, or defining, the scope of disclosure.
    [0006] EXAMPLES [0063] The methods, systems, and tools described herein increase the ability of the beamformer to detect and locate a signal having a frequency much higher than its spatial folding frequency limited by physical network geometry. . The results from the simulations using this method illustrating the improvement are given in FIGS. 7A-7C, 8A-8C and 9A-9C. [0064] FIGS. 7A-7C compare the accuracy (A) of a stationary sensor array with modeled steerable vector weightings independent of a fixed duration (conventional method), (B) of a removable sensor array with steered vector weightings modeled independent of a fixed duration (conventional method), and (C) a removable sensor array with modeled steered vector weightings dependent on an adaptive duration (process of the present application) wherein the source frequency (7 kHz) is less than the spatial fold frequency. FIG. 7B in comparison with FIG. 1A illustrates that moving the network while computing with a steerable fixed vector beamformer which is independent of the duration scrambles the data. FIG. 7C illustrates that the methods, tools and systems of the present disclosure which combine a moving network and a time-dependent steerable adaptive weighting vector beamformer produce an output for the frequency of the acoustic source. commensurable in its accuracy with the conditions of Figure 7A. Since in practice the sensor array would be moving, this example illustrates the accuracy and applicability of the examples of the methods, tools and systems described herein. [0066] FIGS. 8A-8C compare the accuracy (A) of a stationary sensor array with modeled steerable vector weightings independent of a fixed duration, (B) of a removable sensor array with 25 weightings of Model-independent orientable vectors independent of a fixed duration and (C) of a removable sensor array with modeled steered vector weightings dependent on an adaptive duration in which the source frequency (25 kHz) is greater than the frequency of folding spatial. Similar to the example illustrated in FIGS. 7A-7C, the combination of a moving network and a time-dependent directional adaptive weighting vector beamformer produces an output for the frequency of the acoustic source. commensurable in its accuracy with the conditions of Figure 8A. However, in this example, the output of the beamformer shown in FIG. 8B is no longer able to locate the acoustic source. This example illustrates that the exemplary methods, tools, and systems described herein are capable of improving both accuracy and reducing spatial aliasing effects in leak detection. [0068] FIGS. 9A-9C compare the accuracy (A) of a stationary sensor array with modeled steerable vector weightings independent of a fixed duration, (B) of a removable sensor array with weightings of model-independent steered vectors independent of a fixed duration and (C) of a removable sensor array with modeled steered vector weightings dependent on an adaptive duration in which the source frequency (47 kHz) is lower than the frequency of the spatial folding. In FIG. 9A, the output of the resulting beamformer contains much more ambiguity because of many more folds compared to FIG. 8A, which can also be seen in a comparison of FIGS. 8B and 9B. . However, under the conditions of FIG. 9C, the output of the illustrated beamformer demonstrates that the spatial folding effect is reduced and the acoustic signal can be easily located. [0070] Therefore, the present invention is well suited to achieve the ends and obtain the advantages mentioned herein as well as those that are inherent in the present description. The particular embodiments disclosed above are illustrative only, and the teachings of the present disclosure may be modified and practiced in different but equivalent ways that will be apparent to a subject matter expert who benefits from these teachings. In addition, no limitation is provided to the construction or design details described herein, other than those described in the claims below. It is therefore obvious that the given illustrative embodiments disclosed above may be altered, combined or modified and all such variations are considered within the scope and spirit of the present disclosure. The disclosure, as illustratively appropriate, herein can be conveniently practiced in the absence of any element not specifically disclosed herein and / or any optional element described herein. Although the compositions and methods are described herein in terms of "comprising", "containing" or "including" various components or steps, the compositions and methods may also "consist essentially of" or "3037993 24 consist of" various components and steps. All figures and intervals disclosed above may vary by a certain amount. When a numerical range with a lower limit and an upper limit is indicated, any number and range included within the range are specifically indicated. In particular, each range of values (of the form, "from about a to about b" or, equivalently, "from about a to b", or, equivalently, "from about ab") indicated here should be understood as describing each number and interval within the widest range of values. Furthermore, the terms in the claims have their plain and ordinary meaning except in the case of explicit and clear indication other defined by the applicant. In addition, the indefinite articles "a" or "an" as used in the claims are defined herein to mean one or more of the element they introduce.
    权利要求:
    Claims (20)
    [0001]
    REVENDICATIONS1. A method comprising: moving a sensor array comprising at least three sensors along a wellbore; substantially simultaneously measuring signals from an acoustic source with each of the at least three sensors; and processing the acoustic signals in association using a network signal processing technique that uses time-dependent modeled adaptive adaptive vector weights to identify a location of an acoustic signal by a depth and a radial distance from 'a well.
    [0002]
    The method of claim 1, wherein the acoustic sensors form a linear array along a longitudinal axis of the wellbore.
    [0003]
    3. The method of claim 1 or claim 2, wherein the network signal processing technique comprises at least one selected from the group consisting of spatial filtering, delay time estimation. and a sound-energy method.
    [0004]
    The method of claim 1 or claim 2, wherein the acoustic sensors comprise one selected from the group consisting of: an omnidirectional hydrophone, a Bragg fiber grating sensor or a cable of optical fiber, and any combination thereof;
    [0005]
    The method of claim 1 or claim 2, further comprising calculating an azimuthal position of the acoustic source.
    [0006]
    The method of claim 1 or claim 2, wherein the acoustic signals have a frequency higher than a spatial folding frequency used in the network signal processing technique.
    [0007]
    The method of claim 1 or claim 2, wherein the acoustic signals have a frequency lower than a spatial folding frequency used in the network signal processing technique. 30
    [0008]
    The method of claim 1 or 2, wherein a stationary condition 0.011r.1 of the sensor array is preserved by satisfying ATn <where ATn is a duration of the acoustic signal, rj is a location of the first sensor of the at least one Three sensors, and V is a speed of the first sensor moving along the wellbore.
    [0009]
    The method of claim 8, wherein the network signal processing technique comprises: applying a truncation operation to the acoustic signals to produce frames; applying a covariance matrix sample estimator to the frames to produce a covariance matrix sample; the application of a Capon Spatial Spatial Construct operation using the weighted, time-weighted, model-independent orientable vector weights of the covariance matrix sample to produce a Capon Spatial Spectrum Response; and identifying the location of the acoustic signal based on the response of the Capon Spatial Spectrum.
    [0010]
    The method of claim 1 or claim 2, wherein the network signal processing technique comprises: applying a frame decomposition operation to the acoustic signals to produce decomposed frames; applying a covariance matrix sample estimator to the decomposed frames to produce a covariance matrix sample; applying a Capon Spatial Spatial Construct operation using model dependent adaptive steerable vectors dependent on the sample duration of the covariance matrix to produce a Capon spatial spectrum response; and applying a normalization and addition operation to the Capon Spatial Spectrum response to produce a weighted average response of the Capon Spectrum; and identifying the location of the acoustic signal based on the weighted average response of the Capon spatial spectrum. 30
    [0011]
    A system comprising: a sensor array movable along a wellbore and including at least three acoustic sensors for substantially simultaneously measuring and receiving acoustic signals from an acoustic source; ; and a computer-readable non-transitory medium encoded by instructions which, when executed, perform the operations comprising: substantially simultaneously measuring signals from an acoustic source with each of the at least three sensors in a network sensors moving along a wellbore; and acoustic signal processing in combination using a network signal processing technique which comprises a time dependent, scalable adaptive vector for determining a depth and a radial distance from a wellbore of the acoustic source.
    [0012]
    The system of claim 11, wherein the acoustic sensors form a linear array along a longitudinal axis of the wellbore.
    [0013]
    The system of claim 11 or claim 12, wherein the network signal processing technique comprises at least one selected from a group of spatial filtering, an estimate of a delay time and a sound method. -energy.
    [0014]
    The system of claim 11 or claim 12, wherein the acoustic sensors comprise one selected from the group consisting of: an omnidirectional hydrophone, a Bragg fiber grating sensor or an optical fiber cable, and any combination of these;
    [0015]
    A computer-readable non-transitory medium encoded by instructions which, when executed, perform the operations including: substantially simultaneously measuring signals from an acoustic source with each of the at least three sensors in a network sensors moving along a wellbore; and acoustic signal processing in combination using a network signal processing technique which comprises a time dependent, modeled adaptive adaptive vector for determining a depth and a radial distance from a wellbore of the acoustic source.
    [0016]
    The computer-readable non-transitory medium of claim 15 or claim 16, wherein the acoustic sensors form a linear array along a longitudinal axis of the wellbore. 3037993 28
    [0017]
    The computer-readable non-transitory medium of claim 15 or claim 16, wherein the instructions, which when executed, perform the operations which also include: calculating an azimuthal position of the acoustic source. 5
    [0018]
    The computer-readable non-transitory medium of claim 15 or claim 16, wherein a stationary condition of the sensor array is conserved by satisfying ATn <, where ATn is a duration of the acoustic signal, vz ri is a location. a first sensor of at least three sensors, and Vz is the speed at which the first sensor moves along the wellbore. 10
    [0019]
    The computer-readable non-transitory medium of claim 15 or claim 16, wherein the network signal processing technique comprises: applying a truncation operation to the acoustic signals to produce frames; Applying a covariance matrix sample estimator to the frames to produce a covariance matrix sample; applying a Capon Spatial Spatial Construct operation using the weighted, time-weighted, model-independent orientable vector weights to the covariance matrix sample to produce a Capon Spatial Spectrum Response; and identifying the location of the acoustic signal based on the response of the Capon Spatial Spectrum.
    [0020]
    The computer-readable non-transitory medium of claim 15 or claim 16, wherein the network signal processing technique comprises: applying a raster decomposition operation to the acoustic signals to produce decomposed frames; applying a covariance matrix sample estimator to the decomposed frames to produce a covariance matrix sample; The application of a Capon Spatial Spatial Construct operation using the modeled adaptive steerable vector weights dependent on the sample duration of the covariance matrix to produce a Capon Spatial Spectrum Response; and applying a normalization and addition operation to the Capon Spatial Spectrum Response to produce a weighted average response of the Capon Spectrum; and identifying the location of the acoustic signal based on the weighted average response of the Capon spatial spectrum.
    类似技术:
    公开号 | 公开日 | 专利标题
    FR3037993A1|2016-12-30|
    Daley et al.2016|Field testing of modular borehole monitoring with simultaneous distributed acoustic sensing and geophone vertical seismic profiles at Citronelle, Alabama
    EP3164742B1|2021-11-03|Noise removal for distributed acoustic sensing data
    Mateeva et al.2014|Distributed acoustic sensing for reservoir monitoring with vertical seismic profiling
    US9719846B2|2017-08-01|Mud pulse detection using distributed acoustic sensing
    EP0176408B1|1989-05-17|Method and measuring apparatus for the inspection of a borehole provided with a tube
    US10401519B2|2019-09-03|Noise removal for distributed acoustic sensing data
    FR2855272A1|2004-11-26|Characterizing underground formation comprises use of diagraph at bottom of borehole which is rotated about azimuth to allow directional measurements to be taken
    EP2253970B1|2014-01-22|Method for imaging a target area of the subsoil using walkaway data
    FR3036728A1|2016-12-02|
    FR3068479A1|2019-01-04|ANGULAR RESPONSE COMPENSATION FOR VSP DSP
    WO2012038647A1|2012-03-29|Method for estimating elastic parameters by inverting 4d seismic measurements
    FR3069930A1|2019-02-08|ESTIMATING THE SPEED OF TRAINING BY VERTICAL SEISMIC PROFILING
    FR3067745A1|2018-12-21|IMPROVED WAVEFORM ANALYSIS FOR TARGET BOREHOLE WAVE MODES
    WO2012010790A2|2012-01-26|Method for estimating elastic parameters through inversion of 4d seismic measures
    FR3049355A1|2017-09-29|
    FR3077142A1|2019-07-26|OPTIMIZING A REFERENCE LENGTH FOR SIGNAL PRESERVATION AND PROCESSING A REFERENCE LENGTH FOR DISTRIBUTED VIBRATION DETECTION PURPOSES
    Götz et al.2015|Using a fibre optic cable as Distributed Acoustic Sensor for Vertical Seismic Profiling at the Ketzin CO2 storage site
    FR3057606A1|2018-04-20|SYSTEMS AND METHODS FOR USING A SENSOR TO PROVIDE SPATIAL RESOLUTION IN DETECTION OF WELLBORE LEAKS
    EP0944847B1|2002-09-04|Method for seismic processing and in particular for three-dimensional seismic exploration using seismic data migration
    FR3037144A1|2016-12-09|
    FR3061298A1|2018-06-29|ELECTRO-MAGNETIC DOWNHOLE CORROSION DETECTION USING KNOWN PIPE ELEMENTS AND REJECTION OF NATURAL PIPE VARIATIONS
    FR3061976A1|2018-07-20|SEISMIC WELL LINKS USING LOCKOUT SCHEMES
    EP1398649A1|2004-03-17|Method for setting a wellbore
    Jestin et al.2020|On the impact of the gauge length value on Distributed Acoustic Sensing data quality
    同族专利:
    公开号 | 公开日
    MX2017013034A|2017-12-08|
    BR112017021695A2|2018-07-10|
    GB2555280B|2021-03-03|
    GB2555280A|2018-04-25|
    US20170184751A1|2017-06-29|
    US10520626B2|2019-12-31|
    DE112016001828T5|2018-01-04|
    WO2016209388A1|2016-12-29|
    GB201719285D0|2018-01-03|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US4114721A|1977-02-28|1978-09-19|Mobil Oil Corporation|Method and system for acoustic noise logging|
    US5917160A|1994-08-31|1999-06-29|Exxon Production Research Company|Single well system for mapping sources of acoustic energy|
    GB2337591B|1998-05-20|2000-07-12|Geco As|Adaptive seismic noise and interference attenuation method|
    EP1484473B1|2003-06-06|2005-08-24|Services Petroliers Schlumberger|Method and apparatus for acoustic detection of a fluid leak behind a casing of a borehole|
    US8041510B2|2005-11-03|2011-10-18|Saudi Arabian Oil Company|Continuous reservoir monitoring for fluid pathways using microseismic data|
    US7894300B2|2007-01-18|2011-02-22|Schlumberger Technology Corporation|Fluid characterization from acoustic logging data|
    BRPI0817874A2|2007-10-10|2015-03-31|Tecwel As|Method and system for recording and measuring leaks and flows|
    US8345509B2|2009-04-16|2013-01-01|Chevron U.S.A., Inc.|System and method to create three-dimensional images of non-linear acoustic properties in a region remote from a borehole|
    WO2011022408A1|2009-08-18|2011-02-24|Halliburton Energy Services Inc.|Apparatus and method for determining formation anisotropy|
    EP2459847A2|2010-06-10|2012-06-06|Hipoint Reservoir Imaging|Reservoir mapping with fracture pulse signal|
    US8811114B2|2010-08-23|2014-08-19|Baker Hughes Incorporated|Imaging of formation structure ahead of the drill-bit|
    EP2637693B1|2010-11-12|2019-01-16|Chevron U.S.A., Inc.|System and method for generating micro-seismic events and characterizing properties of a medium with non-linear acoustic interactions|
    US9611730B2|2012-04-03|2017-04-04|Weatherford Technology Holdings, Llc|Manipulation of multi-component geophone data to identify downhole conditions|
    US20140104979A1|2012-08-09|2014-04-17|Stolar Research Corporation|Ground-Penetrating Tunnel-Detecting Active Sonar|
    US9523784B2|2012-12-18|2016-12-20|Schlumberger Technology Corporation|Data processing systems and methods for downhole seismic investigations|
    EP3033488A1|2013-10-03|2016-06-22|Halliburton Energy Services, Inc.|Downhole measurement and survey tools with conformable sensors|
    EP3069171B1|2013-11-12|2021-08-25|Westerngeco LLC|Seismic data processing|
    US9798030B2|2013-12-23|2017-10-24|General Electric Company|Subsea equipment acoustic monitoring system|
    US9951585B1|2014-01-30|2018-04-24|William W. Volk|Method of inducing micro-seismic fractures and dislocations of fractures|
    US20160084062A1|2014-09-18|2016-03-24|Sercel|Apparatus and method for a retrievable semi-permanent monitoring system|
    US10222506B2|2014-09-25|2019-03-05|Schlumberger Technology Corporation|System and methodology for orientation of a borehole seismic source|
    US9995837B2|2014-11-07|2018-06-12|Halliburton Energy Services, Inc.|Apparatus and methods of extracting reflections from acoustic array data|
    WO2016115012A1|2015-01-13|2016-07-21|Halliburton Energy Services, Inc.|Acoustic array signal processing for flow detection|WO2016085769A1|2014-11-24|2016-06-02|Sikorsky Aircraft Corporation|Multispectral sensor fusion system for platform state estimation|
    BR112019018817A2|2017-05-25|2020-04-14|Halliburton Energy Services Inc|method to detect well bottom leak and system to detect well bottom leak|
    US20200271808A1|2017-08-23|2020-08-27|Hallburton Energy Services, Inc.|Synthetic aperture to image leaks and sound sources|
    WO2019132975A1|2017-12-29|2019-07-04|Halliburton Energy Services, Inc.|Systems and methods to utilize sensors to provide spatial rosolution in downhole leak detection|
    US10371797B1|2018-05-23|2019-08-06|Zendar Inc.|Systems and methods for enhancing target detection|
    WO2020005281A1|2018-06-29|2020-01-02|Halliburton Energy Services, Inc.|Systems and methods to utilize a sensor to provide spatial resolution in downhole leak detection|
    CN111879970A|2020-08-31|2020-11-03|防灾科技学院|Temperature insensitive FBG acceleration sensor and method based on strain chirp effect|
    法律状态:
    2017-04-12| PLFP| Fee payment|Year of fee payment: 2 |
    优先权:
    申请号 | 申请日 | 专利标题
    US201562184995P| true| 2015-06-26|2015-06-26|
    PCT/US2016/031772|WO2016209388A1|2015-06-26|2016-05-11|Continuous beamforming while moving: method to reduce spatial aliasing in leak detection|
    [返回顶部]