JPH04203196A - Underground structure evaluation method based on analysis of three dimensional particle movement of discharge elastic wave during digging of chute - Google Patents

Abstract

PURPOSE:To facilitate analysis of underground structure by detecting elastic wave which is discharged from drill bit by an AE sensor in a chute, determining the direction of digging bit via three dimensional analysis, and learning the existence of reflection surface based on difference of time between direct wave and reflected wave. CONSTITUTION:Blastic wave which is discharged from a drill bit 11 during digging is detected by an AE sensor 14 for three axis AE measurement which is provided in a monitoring chute 15 which is digged apart from a digging chute 12, and AE signal is sampled by a measuring car 16. Next, AE signal is filtered in the frequency range which is sufficiently lower than resonance frequency of the sensor 14, and the direction of the bit 11 is detected by the analysis of three dimensional particle movement. Reflection surface of elastic wave which exists in the direction of digging (boundary of ground layers) is evaluated based on the direction of relative progress and difference in reach time of direct wave and reflection wave. Consequently, it is possible to perform the analysis of underground structure easily at real time.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention provides that the so-called Lissajous pattern, which is a waveform representing the three-dimensional particle motion of elastic waves (acoustic emissions) generated during drilling of a well, is applied to a drilling bit. Based on the discovery of the fact that the direction of the drilling bit is oriented in the direction of Predict the existence of surfaces (fracture zones, cracks, strata boundaries, etc.) in advance, and (2) understand the underground structure around the wellbore (vs.
The present invention relates to an underground structure evaluation method using three-dimensional particle motion analysis of emitted elastic waves during well drilling, which makes possible p).

[Prior Art] Conventionally, as means for detecting underground structures using a three-axis detector using sound waves, etc., methods described in JP-A No. 32085 of 1983, JP-A No. 90495 of 1982, and There is one described in Japanese Patent Application Laid-Open No. 59792 of 1988. In addition, so-called seismic wave data such as vibration waves or elastic waves generated by the excavation bit during excavation is acquired through a geometer installed on the ground surface. A technique called the TOMEX method is known in which the vibration propagation time of the data obtained between these two points is corrected, and the underground structure is analyzed based on this.

This conventional technique called the TOMEX method, in addition to the basic configuration of a drilling bit 1, wellbore 2, and drilling rig 3, as shown in FIG.
installed on the ground surface away from the drilling bit 1, and captures the direct waves and reflected waves of the vibration waves generated when the drilling bit 1 excavates at the drilling rig 3 position and the geoffon 4 installation position as reference points, respectively, By taking the mutual correlation between them, for example, a stratum boundary existing in the vicinity of the well 2 can be detected. That is, the vibration propagation time between the excavation bit 1 and the reference point is corrected for the observation data at the reference point to obtain a vibration wave meter, that is, input wave data from the excavation bit 1 position. Then, a cross-correlation process is performed on this input wave data and the data obtained by the geofone 4 on the ground surface to obtain a record similar to that of an impulse epicenter.

[Problem that the invention seeks to solve]

This type of TOMEX method calculates the difference in propagation distance between the direct wave from the tip of the drilling bit and the reflected wave from the reflecting surface to the detector, and calculates the correlation between the reference signal from the drilling rig and the detected signal. The reflective surface is estimated by analysis, but the analysis time usually takes about 1 hour.
It is necessary to analyze signals over a long period of time,
Not only is this method unsuitable as a real-time monitoring method desired in the field, but it also has the disadvantage that analysis itself becomes impossible when the rate of well drilling is large.

In addition, in the conventional TOMEX method, since only the reflected waves are used for inclined reflecting surfaces, it is difficult to estimate the reflected waves, and when estimation is absolutely necessary, This method requires the introduction of a separate sensor array and migration analysis, which not only results in high costs but also has the disadvantage of increasing the amount of analysis and reducing accuracy.

Furthermore, in this method, it is necessary to estimate the sound emitted from the tip of the drilling bit using signals detected in the rig, and therefore it is necessary to estimate the transfer function of the drilling system with high accuracy.

Furthermore, this method is strongly influenced by noise propagating on the ground surface and high attenuation layers near the ground surface, so there is a decisive drawback in that the resolution deteriorates due to noise and loss of one high frequency component.

[Means to solve the above problems]

In order to solve the above-mentioned drawbacks, the inventors of the present application conducted in-well triaxial acoustic emission measurement (AE measurement) during well drilling, and found that the three-dimensional Lissajous shape obtained by this measurement was In other words, the direction of polarization obtained from the principal axis analysis of the surge pattern in a limited frequency band is extremely consistent with the direction of the drill bit in the wellbore. This is because in the AE signal, the directly arriving longitudinal wave component is more prominent than the reflected wave or mode-converted transverse wave, and the Lissajous pattern shows extremely good directionality.

Therefore, by analyzing this Lissajous pattern, it is possible to detect the direction of the drilling bit extremely easily and with a simple configuration.Furthermore, this information can be used to detect weak reflected waves from strata boundaries. By detecting this and evaluating its delay time, it is possible to detect strata boundaries that exist at higher depths in the direction of progress during excavation.

[Effect]

The present invention utilizes a well drilling drill bit as an AE signal source and performs three-dimensional particle motion analysis on this AE multiplied signal obtained by a triaxial AE sensor.
For a specific frequency, the polarization direction obtained from the principal axis analysis of the Lissajous pattern matches the direction of the actual drill bit, so the direction of the drill bit can be determined by this analysis, and the result can be used to determine the direction of the drill bit. , the stratum boundaries existing in the direction of excavation progress are evaluated from the relative travel directions and arrival time differences of direct waves and reflected waves.

〔Example〕

The present invention will be explained based on a conceptual diagram of an embodiment. FIG. 1 is a conceptual diagram of AE triaxial measurement according to the present invention.

In the figure, 11 is a drill bit, 12 is a drilling well, 1
3 is a drilling rig, 14 is a measurement well for monitoring, and 15 is a triaxial AE
The sensor 16 is a measuring vehicle. Now, in order to understand the concept of this system, a measurement outline of the measurement vehicle 16 is shown in the figure. That is, in the figure, 17 is a three-axis main amplifier (3Ch, Main Amp), and 18 is a fixing motor controller (Fixing MotorCon).
19 is an electronic orientation detection device (El
electric Compass Driver), 2
0 is an automatic data collection system (Automatic d
ata Acquisoon System), 21
is a personal computer, 22 is a recording disk, 23
is a data recorder, 24 is a digital storage oscilloscope, 25 is an RMS DC converter, and 26 is an X-recorder. For the purpose of drilling a well of 380 m based on a system with such a configuration, 313 m - 355 m
The drill bit 11 is rotated between the holes to generate the sound of rock crushing due to drilling, and the sound is transmitted to the triaxial AE measurement sensor installed in the monitoring well 15 drilled at a distance of about 80 m from the drill well 12. AE-fold signal was collected by 14. This in-well AE sensor 14 was installed in the monitoring well 15 at a depth of 206 m.

In such a measurement system, as a result of performing three-dimensional particle motion analysis on the AE multiplied signal obtained by the three-axis AE sensor 14, the surge pattern in the raw state is shown in FIG. Figure 3 shows the results of filtering at the cutoff frequency.

In Figures 2 and 3, Figure (a) shows the horizontal glue surge pattern, and Figure (b) shows the vertical glue surge pattern, with the drill bit 11 in the direction of the arrow. shall be. If we compare and examine the Lissajous patterns shown in Figures 2 and 3 of 2, we can see that the raw data glue surge pattern (Figure 2) does not show much polarization, but the glue surge pattern as a result of filtering has an elliptical shape. (FIG. 3) shows that the main axis direction corresponds to the direction of the actual drill bit 11.

In addition, spectral matrix analysis of AE data [J, C, Sa
msom,”Matrix and 5tokes V
ectorRepresentation of De
Tectors for Polariz-ed Wa
veforms: Theory and Some A
Applications to Teleseimi
c Waves,” Geophycs, J, RAst
r,Soc, ,51. pp, 583-603 (197
7)] The vertical projection of the eigenvector obtained by
- As an example, as shown in FIG. This figure shows the polarization direction as a function of frequency, and as is clear from this figure, the AE Lissajous pattern in the frequency range of 100 Hz or less, which is sufficiently lower than the resonant frequency of the AE sensor 10, is similar to that of the drill bit. This shows that it is polarized in the direction. This seems to be due to the following reasons.

(a) Among the elastic waves generated by the drilling bit, longitudinal waves are more dominant than transverse waves.

(b) Reflection, refraction, and mode conversion waves of elastic waves at strata boundaries are relatively small.

Therefore, in a frequency range sufficiently lower than the resonant frequency of the sensor, the direction of the drill bit can be determined by AE multiplied signal three-dimensional Lissajous analysis.

Based on the above results, the inventors conducted principal axis component analysis (PCA) of the AE Lissajous pattern through a low-pass filter of fc = 80 Hz to detect the drilling bit direction.
I did this. The results are shown in FIG. FIG. 5(a) shows the results of analyzing the horizontal component, and FIG. 5(b) shows the results of analyzing the vertical component. As is clear from FIGS. 5(a) and 5(b), the direction of the excavation bit can be detected with extremely high accuracy using PCA in both the horizontal and vertical directions. In this example, as a result of comparing and contrasting with the actual drilling bit direction,
The detection error was within 15 degrees in both horizontal and vertical image planes.

This error is thought to be due to the underground velocity structure and added noise.

In addition, in this example, it was found that strong energy waves generated by friction between the drill vibe and the pit wall sometimes, although not always, affect the detection of the position of the drilling bit. That is, the strength of the friction is included in the RMS (effective value) signal of AE and is generated as a constant periodic signal corresponding to the rotation of the drill pipe, so it can be evaluated by the energy density distribution ratio (EDDR'). can. Figure 6 shows the relationship between EDDR and the angular error of the drilling bit position. In Figure 6, the orientation error is EDDR
It can be seen that ' tends to increase in a high state of 10 or more.

Based on the above results, the inventors of the present application decided to apply not only the direct wave from the drill bit 11 but also the reflected wave to the evaluation of underground structures. In other words, as mentioned above, AE caused by rock fracture due to excavation
Since longitudinal waves are more dominant than double signals and transverse waves, this A
By detecting the E-fold signal direct wave and reflected wave, underground structure analysis becomes possible.

Therefore, the inventors of the present application developed a method shown in FIG.
Spit measurement was performed. FIG. 7 shows only the measurement concept, and its basic measurement configuration is not different from that shown in FIG. 1. Therefore, the same components as those used in FIG. 1 are designated by the same reference numerals. In addition, 30 is a stratum boundary shown conceptually. As shown in FIG. 7, this stratum boundary can be evaluated from both the relative wave traveling direction and time delay of the direct wave and the reflected wave.

In other words, in order to estimate the cross-spectral analysis of particle motion as an evaluation of the arrival time difference, it is necessary to calculate the following: ``(a) The drilling bit generates a longitudinal wave as a point signal source, (b) The energy of the mode conversion to a transverse wave at the strata boundary. The calculation was made on the assumption that the reflection of longitudinal waves is negligibly small compared to the reflection of longitudinal waves. Under such a premise, the vector of particle motion by the three-axis AE sensor can be expressed as follows.

P(t)=Pd(t)+Pr(t)+fL(t)
- (1) Here, Pd(t) and Pr(t) are particle motion vectors of the longitudinal wave that arrived directly and the longitudinal wave that arrived by reflection, respectively. Further, n (t) is a vector of noise components. Assuming that the reflected longitudinal wave has a time delay τ and a difference in the traveling direction in the vertical plane θ, the difference between the time delay and the traveling direction is expressed as the wave particle motion vector Pv(ω , φ). Here, the particle motion vector PV (
ω, φ) can be calculated from the three-dimensional particle motion if the direction of the actual AE signal source is known. That is, the cross spectrum of the direct wave and the virtual reflected wave can be expressed as a function of the angular frequency ω and the relative direction ψ of the reflecting surface, and the following equation is derived.

S (ω, φ)= F d(ω)X F ”v(ω, φ
)・(2) Here, ' represents a conjugate complex number.

The relative direction of the reflective surface can be determined from the relative direction ψ of the virtual reflective surface at the peak of +51(ω,φ)1, and the relative distance to the reflective surface is also determined by the cross spectrum ≦
It can be determined from the phase of (ω, φ).

When the result obtained by this method is applied to the filtered AE multiplied signal, the result becomes as shown in FIG.

As can be seen from Figure 8, the displayed cross spectrum! It can be seen that there are several peaks in the norm of (ω, φ). That is, the reflective surface is estimated in the direction of ψ that gives the maximum value to the norm of the cross spectrum.

Figures 9 and 10 respectively show the variation in distance and direction obtained as a function of excavation depth by selecting the maximum value of the cross spectrum in the frequency range 20-40 Hz.

This change is due to the fact that the reflective layer in this field is 385 m
A silt layer is assumed to exist in the area, and this is shown with a broken line. In FIG. 9, the distance appears to be slightly off, but when compared with FIG. 10, it can be seen that the direction matches the actual silt surface.

In this embodiment, monitoring wells were separately provided in addition to the drilling wells, but the number is not limited to this number. By installing multiple wells, the accuracy of three-dimensional underground analysis will be significantly improved.

〔Effect of the invention〕

The present invention uses the excavation bit during excavation as a signal source, supplements it with a three-axis AE sensor, and performs three-dimensional analysis.The filtered AE multiplied signal surge pattern is polarized in the direction of the excavation bit. The direction of the drilling bit can be determined.

Furthermore, the presence of a reflective surface can be easily determined from the time difference between the direct wave and the reflected wave from the excavation bit, which can contribute to underground structure analysis in the direction of excavation progress.

That is, according to the present invention, the reflection surface can be easily estimated by cross-spectral analysis of particle motion detected by the in-well triaxial AE sensor, so it is possible to estimate the underground structure independently of the excavation work. Moreover, an extremely excellent effect can be found in that underground structures can be analyzed by analyzing signals obtained in just a few minutes.

This means that real-time monitoring is easily possible on-site, and as a result, it is possible to predict the presence of fracture zones, reservoirs, etc. in advance when excavating for underground structure evaluation. effective.

Furthermore, since it is possible to evaluate the underground structure using the propagation distance difference and propagation direction difference between direct waves and reflected waves, it is also possible to estimate the slope underground, This has extremely excellent effects such as enabling highly accurate estimation of the structure.

Furthermore, unlike the conventional TOPEX method, the present invention does not require estimation of the signal emitted from the tip of the drilling bit, so it is possible to reduce the calculation time, and this also allows the signal transmission of the drill vibe. Uncertainty elements such as functions can be reduced.

In addition, since the measurement is performed inside the well, it is possible to avoid the effects of noise and high attenuation layers near the ground surface, so high S/
Since it is possible to perform N measurement and also detect high frequency components, it is possible to sell an extremely excellent underground structure evaluation method that enables analysis with high resolution.

[Brief explanation of the drawing]

FIG. 1 is a conceptual diagram showing an example of AE three-axis measurement according to the present invention, and FIG. 2 is a conceptual diagram showing an example of AE three-axis measurement according to the present invention, and FIG. Figure 3, which shows the glue surge pattern, shows this at 80Hz.
Figure 4 shows an example of the vertical projection of the eigenvector obtained by the spectral matrix analysis of the AE data, Figure 5 (A) and (B). is, f. Figure 5 (a) shows the results of principal axis component analysis (PCA) of the AE Lissajous pattern passed through a low-pass filter of =80Hz.
) shows the analysis results for the horizontal component, (b) shows the analysis results for the vertical component, and Figure 6 shows the energy density distribution ratio (EDDR')
A diagram showing the relationship between and the angular error of the drilling bit position,
Figure 7 shows three-axis VS using reflected waves from a drill bit.
Figure 8, a conceptual diagram of P measurement, shows the particle motion vector Pv(ω,φ
) Figures 9 and 1-0 show the results of applying the filtered AE multiplied signal to the results expressed using the concept of FIG. 11 is a diagram showing the changes in distance and direction obtained as a function of excavation depth, and is a diagram showing the basic concept of the conventional AE three-axis measurement called the TOMEX method. DESCRIPTION OF SYMBOLS 1... Drilling bit, 2... Well, 3... Drilling rig, 4... Geofluorescence, 10... AE sensor, 11... Drill bit, 12... Drilling well, 14 ...In-well AE sensor, 16...Measurement vehicle, P...Particle motion vector, S...Cross spectrum, τ...Time delay, ψ...Relative direction, ω...Angle Frequency-1 Patent applicant President of Tohoku University (1 other person) and agent
9004 Masashi Hitoshi Otaki Figure 6 i: ot (a7 nonzontat Figure 2 @ Figure 3 Figure 7 Section 8 Figure D "Depth (m) Figure 9 Drilling Depth (m) Figure 10

Claims (1)

[Scope of Claims] Claim 1: The elastic waves emitted from the drilling drill bit during drilling are detected by an in-well triaxial elastic wave detector, and the detection results include a frequency range sufficiently lower than the resonant frequency of the sensor. A method of detecting the direction of the drill bit by filtering the AE signal and analyzing the principal axis component of the three-dimensional particle motion of the AE signal. Claim 2: An in-wellbore three-axis elastic wave detector detects elastic waves emitted from a drilling drill bit during drilling, and in cross-spectral analysis of three-dimensional particle motion as a result of the detection, direct waves and reflections from the bit are detected. A method for evaluating underground structures using three-dimensional particle motion analysis of elastic waves emitted during well drilling, which analyzes underground structures from the relative traveling directions of waves and their arrival time differences. Claim 3: An in-well triaxial elastic wave detector detects elastic waves emitted from a drilling drill bit during drilling, and in cross-spectral analysis of three-dimensional particle motion of the detection results, direct waves and reflections from the bit are detected. During well drilling according to claim 2, wherein a stratum boundary existing in the direction of excavation progress is evaluated from both a difference in the transmission distance of the reflected wave from the surface and a difference in the propagation direction of the direct wave and the reflected wave to the sensor. A method for evaluating underground structures using three-dimensional particle motion analysis of emitted elastic waves. Claim 4: The cross-spectral analysis of three-dimensional particle motion detected by the in-well triaxial acoustic wave detector is obtained as a result of filtering at a predetermined cutoff frequency. The drilling drill bit direction detection method described. Claim 5: Claim 2, wherein the cross-spectral analysis of the three-dimensional particle motion detected by the in-well triaxial acoustic wave detector is obtained as a result of filtering at a predetermined cutoff frequency. A method for evaluating an underground structure by three-dimensional particle motion analysis of elastic waves emitted during well drilling according to items 3 to 3. Claim 6: In the cross-spectral analysis of the three-dimensional particle motion detected by the in-well three-axis acoustic wave detector, the vector of particle motion by the three-axis AE sensor is defined as P(t)=Pd(t)+Pr(t)+n. (t)…(1)
(Pd(t) and Pr(t) are particle motion vectors of the longitudinal wave that arrived directly and the longitudinal wave that arrived by reflection, respectively) and the cross spectrum of the direct wave and the virtual reflected wave with the angular frequency ω and the virtual A function of the relative direction φ of the reflecting surface of , S(ω, φ)=Pd(ω)×P^*v(ω, φ)…(2
) (^* is a conjugate complex number) Underground structure evaluation by three-dimensional particle motion analysis of elastic waves emitted during well drilling according to claim 2 or 3, characterized in that processing is performed by the formula expressed as Method.