CN104459797A - Method for recognizing and collecting microseism events in well - Google Patents

Abstract

The invention provides a method for recognizing and collecting microseism events in a well and belongs to the field of microseism monitoring. The method comprises the steps that (1) orientation rotation is conducted on each receiving point recorded in the even monitoring process; (2) for records obtained after rotation is conducted, a characteristic function value CFs of each receiving point at each moment is calculated; (3) an event superposition energy function L(t) is calculated through a time difference scanning technique; (4) the seism starting time ti of events is automatically detected through self-adaptive threshold values; (5) the peak value time of S waves of a set of events is collected and checked, and the jumping time of the S waves is collected and checked; (6) for each receiving point, a characteristic function value CFp is calculated for each time point within a fixed time interval Tspmax before the jumping time of the determined set of S waves; (7) the peak value time of each receiving point of P waves of the set of events is collected and checked, and the jumping time of the P waves is collected and checked.

Description

A kind of method of borehole microseismic event recognition and pickup

Technical field

The invention belongs to micro-seismic monitoring field, be specifically related to a kind of method of borehole microseismic event recognition and pickup.

Background technology

Along with the further aggravation of hydrocarbon resources supply and demand tight relationship, the sight increasing oil and gas production has been turned to tight sand and shale by petroleum exploration domain.The tight sand of China's abundant and shale hydrocarbon resources will become the mainstay of following hydrocarbon resources.But the exploitation of tight sand and shale hydrocarbon resources depends on hydraulic fracturing technology, the object of waterfrac treatment is in order to by pressure change stratal configuration, makes it to be formed with the crack being beneficial to oil-gas migration, in the production process of crack, along with micro-seismic event.Micro-seismic monitoring can provide foundation for the design of waterfrac treatment and fracturing effect, and then for improving rate of oil and gas recovery service.

In well, three-component micro-seismic monitoring is one of major way of current waterfrac treatment micro-seismic monitoring.It is generally by placing three-component geophone group in well, monitors the micro-seismic event produced in fracturing process continuously.The feature of micro-seismic monitoring is: 1. monitoring time is long, general continuous monitoring more than several hours; 2. event density is large, and once monitoring can detect hundreds of event; 3. event signal is weak, and to-noise ratio is low; 4. event seismic phase type is complicated, and existing P ripple also has S ripple etc.

Therefore, current exist working strength large with the method being manually primary means identification and pickup micro-seismic event, and weak signal pickup is inaccurate, and P ripple S ripple seismic phase is difficult to the problem judged, and says in efficiency, can not be real-time provide foundation for hydraulic pressure.For automatic identification and the pickup of micro-seismic event, current method, although identify and picked up partial event, but there is no the seismic phase of decision event, also not to from the P ripple of same focus and S ripple event to effectively matching, this causes difficulty to the subsequent treatment of micro-seismic monitoring to a certain extent.

In sum, it is long to there is the event monitoring time in current micro-seismic monitoring, and event number is large, and signal is weak, manually carries out event recognition and pickup inefficiency, can not the problem of Real-Time Monitoring.

Summary of the invention

The object of the invention is to solve the difficult problem existed in above-mentioned prior art, a kind of method of borehole microseismic event recognition and pickup is provided, for the micro-seismic event that three-component seismometer in well receives, construct new S ripple and the fundamental function of P ripple respectively, devise new S ripple and the identification of P ripple and pick-up method, improve the efficiency and accuracy rate that event are effectively identified and picks up.

The present invention is achieved by the following technical solutions:

A method for borehole microseismic event recognition and pickup, said method comprising the steps of:

(1) carry out orientation rotation to each acceptance point of event monitoring record, make X component and due east E to coincidence, Y-component and positive northern N are to coincidence, and Z component is vertical;

(2) to postrotational record, the fundamental function value CFs in each each moment of acceptance point is calculated;

(3) by moveout scan technique computes event stack power function L (t);

(4) a shake time t of event is automatically detected by adaptive threshold _i;

(5) pick up the S crest value time of one group of event and check the S crest value time; Pickup S ripple take-off time also checks S ripple take-off time;

(6) for each acceptance point, the fixed time interval T before the one group of S ripple take-off time determined _spmaxinterior to each time point calculating fundamental function value CFp;

(7) pick up each acceptance point time to peak of the P ripple of one group of event and check the P crest value time; Pickup P ripple take-off time also checks P ripple take-off time;

(8) check P ripple, the S ripple match condition of one group of event of having picked up, if ineligible, then reject this event; Check the true and false situation of one group of event of having picked up, if ineligible, then reject this event;

(9) judge whether that all events have all been picked up, if not, then return step (5), if so, then export peak time and the take-off time of all qualified S ripples and P ripple, terminate pick process.

Described step (1) comprising:

(11) establish three of three-component seismometer record components be respectively x ', y ', z ', first use formula (1) by wave detector choosing install to UVW coordinate system under:

$\left[\begin{array}{c}u\\ v\\ w\end{array}\right]=\left[\begin{array}{ccc}\cos \mathrm{\θ}& -\sin \mathrm{\θ}& 0\\ \sin \mathrm{\θ}& \cos \mathrm{\θ}& 0\\ 0& 0& 1\end{array}\right]\left[\begin{array}{c}{x}^{\′}\\ y^{\′}\\ z^{\′}\end{array}\right]---\left(1\right)$

Here θ represents the angle of u and x ';

(12) formula (2) is utilized to continue to rotate u, v, w to ENZ rectangular coordinate system:

$\left[\begin{array}{c}x\\ y\\ z\end{array}\right]=\left[\begin{array}{ccc}{e}_1& e_2& e_3\\ f_1& f_2& f_3\\ g_1& g_2& g_3\end{array}\right]\left[\begin{array}{c}u\\ v\\ w\end{array}\right]---\left(2\right)$

(e in formula ₁, f ₁, g ₁), (e ₂, f ₂, g ₂), (e ₃, f ₃, g ₃) be the direction cosine of UVW tri-axles in ENZ coordinate system respectively;

Wherein $(e_3,f_3,g_3)=(\frac{{\mathrm{\Δx}}_1}{d_1},\frac{{\mathrm{\Δy}}_1}{d_1},\frac{{\mathrm{\Δz}}_1}{d_1})(d_1=\sqrt{{\left({\mathrm{\Δx}}_1\right)}^2+{\left({\mathrm{\Δy}}_1\right)}^2+{\left({\mathrm{\Δz}}_1\right)}^2}),$ Δ x ₁, Δ y ₁, Δ z ₁the increment of coordinate of adjacent wave detector in ENZ coordinate system, by UVW orthogonality relation and derive:

$(e_1,f_1,g_1)=(-\frac{f_3}{\sqrt{1-g_3^2}},\frac{e_3}{\sqrt{1-g_3^2}},0)(e_2,f_2,g_2)=(\frac{e_3{g}_3}{\sqrt{1-g_3^2}},\frac{f_3{g}_3}{\sqrt{1-g_3^2}},\sqrt{1-g_3^2})$

(13) formula (3) is utilized to be threaded to by wave detector in the rectangular coordinate system of shooting point and acceptance point line direction formation, if postrotational three vectors are (a _x, a _y, _az), wherein a _xfor the horizontal direction of line examined by big gun, a _yfor perpendicular to a _xanother transverse axis, a _zvertical is downward:

$\left[\begin{array}{c}{a}_x\\ a_y\\ a_z\end{array}\right]={\left[\begin{array}{ccc}{b}_{11}& b_{12}& 0\\ b_{21}& b_{22}& 0\\ 0& 0& 1\end{array}\right]}^{-1}\left[\begin{array}{c}x\\ y\\ z\end{array}\right]---\left(3\right)$

Make Δ x ₂=x _r-x _s, Δ y ₂=y _r-y _s, Δ z ₂=z _r-z _s; wherein (x _s, y _s, z _s) and (x _r, y _r, z _r) represent shooting point and the coordinate of acceptance point in ENZ coordinate system respectively; Then b ₂₁=b ₁₂, b ₂₂=-b ₁₁;

(14) formula (4) is utilized to continue rotating vector (a _x, a _y, a _z), make a _xwith due east E to coincidence, a _ywith positive northern N to coincidence, a _zstill downwards vertical;

$\left[\begin{array}{c}X\\ Y\\ Z\end{array}\right]=\left[\begin{array}{ccc}\cos \mathrm{\β}& -\sin \mathrm{\β}& 0\\ \cos \mathrm{\β}& \cos \mathrm{\β}& 0\\ 0& 0& 1\end{array}\right]\left[\begin{array}{c}{a}_x\\ a_y\\ a_z\end{array}\right]---\left(4\right)$

Wherein β is a _xwith the angle of E, cos β=Δ x ₂/ d ₂, sin β=Δ y ₂/ d ₂.

(15) method by carrying out angle scanning to the component direct P ripple of perforation data tries to achieve wave detector horizontal azimuth: as postrotational component a _xwith a _ythe maximum angle θ of energy difference be required position angle:

$E\left(\mathrm{\θ}\right)=\underset{k=n1}{\overset{n2}{\mathrm{\Σ}}}{a_x}^2\left(k\right)-\underset{k=n1}{\overset{n2}{\mathrm{\Σ}}}{y_x}^2\left(k\right)---\left(5\right).$

Described step (2) is achieved in that

(21) in a sliding window (number of samples is N, time window size be the one-period of ripple), for these three components of postrotational X, Y, Z, its covariance matrix is built:

$M=\left[\begin{array}{ccc}\mathrm{\ΣAA}& \mathrm{\ΣAB}& \mathrm{\ΣAC}\\ \mathrm{\ΣBA}& \mathrm{\ΣBB}& \mathrm{\ΣBC}\\ \mathrm{\ΣCA}& \mathrm{\ΣCB}& \mathrm{\ΣCC}\end{array}\right]$

(this matrix is existing, its eigenwert and proper vector to solve also be basic mathematical method)

In formula, ∑ is $A=(x_i-\stackrel{\‾}{x}),B=(y_i-\stackrel{\‾}{y}),C=(z_i-\stackrel{\‾}{z}),$ Wherein $\stackrel{\‾}{x}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{x}_i,$ $\stackrel{\‾}{y}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{y}_i,\stackrel{\‾}{z}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{z}_i;$ Asked by covariance matrix and obtain three eigenvalue λ ₁, λ ₂, λ ₃and correspond to three stack features vectors of eigenwert, wherein λ ₁> λ ₂> λ ₃, eigenvalue of maximum λ ₁characteristic of correspondence vector is V1 (v _x, v _y, v _z);

(22) utilize with the eigenvalue of maximum λ 1 of covariance matrix and its characteristic of correspondence vector V1 (v _x, v _y, v _z), and perforation grid bearing vector S (s _x, s _y, s _z) structure S wave characteristic function:

$\mathrm{CFs}=(1-|\frac{s_x{v}_x+s_y{v}_y+s_z{v}_z}{\sqrt{{s_x}^2+{s_y}^2+{s_z}^2}+\sqrt{{v_x}^2+{v_y}^2+{v_z}^2}}\left|\right)\mathrm{\λ}1---\left(6\right)$

Wherein: S (s _x, s _y, s _z)=(x _s-x _r, y _s-y _r, z _s-z _r)

Then formula (6) is utilized to calculate the fundamental function value CFs in each each moment of acceptance point to postrotational record (X, Y, Z).

Described step (3) is achieved in that

Formula (8) is utilized to calculate maximum stack energy L (t) in each moment:

$L\left(t\right)=\underset{j}{\mathrm{\Σ}}{\mathrm{CFs}}_j(t-{\mathrm{\τ}}_j)---\left(8\right)$

In formula (8), τ _jit is the time shift amount of a jth wave detector, this time shift amount can on a 3D grid near shooting point, calculate focus on each grid node by scanning and, to the normal moveout correction time difference of acceptance point, choose the one group of time difference value making L (t) maximum, as the time shift amount τ of current time _j, and the grid node locations making L (t) maximum is exactly the position of doubtful focus, corresponding moment t _i(i refers to i-th group of event) is exactly a shake time of doubtful focus.

Described step (4) comprising:

(41) Hilbert transform is carried out to L (t), calculate envelope H (t) of L (t);

(42) in a sliding window, expectation value E (t) and standard variance δ (t) of each moment envelope H (t) is calculated;

(43) formula (9) is utilized to carry out the calculating of adaptive threshold:

Ht(t)=E(t-τ)+αδ(t-τ) (9)

In formula (9), τ is time delay, and the threshold values caused because of first arrival in order to adjustment changes too early.α is the weight coefficient of standard variance.

(44) at a regular time length T _eventin, the maximal value L that L (t) value is greater than Ht (t) will be met _max(t) as the event in this detection window, corresponding time t _ibe exactly that event plays the shake time, i=1,2,3...K, K are the event numbers detected, described T _eventbe greater than the maximum time difference of S-P ripple; The length of described detection window is time span T _event.

The pickup S crest value time in described step (5) also checks that the S crest value time is achieved in that

The pickup S crest value time: the S crest value time refers to time corresponding to the ceiling capacity of event, to a certain event that step (4) detects, to each acceptance point, with t _i-τ _jmoment is this acceptance point time to peak searching window mid point, searches for the maximal value of CFs at this time in window, and its corresponding time is exactly the time to peak t of event at this acceptance point _s0; t _ia shake time of i-th group of event that step (4) detects, τ _jit is the time shift amount of the jth wave detector calculated in step (3);

Check the S crest value time: (Parabolic Fit is the basic approximating method of one of discrete points data, is the time to peak matching second-degree parabola equation Tfit with pickup with parabolic equation matching time to peak T-X curve here _i=ax _i ²+ bx _i+ c, x _ifor receiving period, Tfit _itime for matching), for error, beyond the mark (difference between match value and actual value is error e rr=|Tfit-Tpick|, Tpick is actual pickup value, wherein the boundary value of err is generally several sampled point time, the error precision of this value required for reality, by pickup effect test determine) time to peak value, with the theoretical peak time Tfit with Parabolic Fit _ifor time window mid point, reduce search window, again pick up the maximal value of CFs, namely in the search window reset, find out the maximal value of CFs;

Pickup S ripple take-off time in described step (5) also checks that S ripple take-off time is achieved in that

Pickup S ripple take-off time: S ripple take-off time refers to the time that the first arrival ski-jump of event is corresponding; After the time to peak of event is determined, with time to peak t _s0for in the take-off time detection window of mid point, the long short time-window slided at two respectively calculates mean value STA and LTA of CFs, then utilizes the maximal value of STA/LTA as the take-off time t of S ripple _s; STA/LTA window can not be overlapping, and simultaneously in order to eliminate the pickup of insecure time, the ratio of STA/LTA is provided with minimum threshold values;

Check S ripple take-off time: with Parabolic Fit take-off time T-X curve, for the take-off time value that error is beyond the mark, with the theoretical take-off time Tfit with Parabolic Fit _ifor time window mid point, reduce search window, reset the STA/LTA threshold values of a local, again pick up the maximum value of STA/LTA, namely in the search window reset, again find out the maximal value of STA/LTA.

Described step (6) is achieved in that

For an acceptance point, the fixed time interval T before the S ripple take-off time determined _spmax(its size and the T with step (44) _eventwith) in, utilize the S wave polarization vector P (p detected _x, p _y, p _z) (P (p _x, p _y, p _z) be the eigenvalue of maximum character pair vector of the covariance matrix of the S crest value time point of pickup in step (5), the V1 (v of the S crest value time namely calculated by step (22) _x, v _y, v _z)), and the eigenvalue of maximum λ 1 of the covariance matrix at time point place to be calculated and character pair vector V1 (v _x, v _y, v _z) structure P wave characteristic function:

$\mathrm{CFp}=(1-|\frac{p_x{v}_x+p_y{v}_y+p_z{v}_z}{\sqrt{{p_x}^2+{p_y}^2+{p_z}^2}+\sqrt{{v_x}^2+{v_y}^2+{v_z}^2}}\left|\right)\mathrm{\λ}1---\left(7\right)$

Then formula (7) is utilized to calculate CFp to each time point of each acceptance point.

The pickup P crest value time in described step (7) also checks that the P crest value time is achieved in that

The pickup P crest value time: at whole time interval T _spmaxin, the maximal value of search CFp, the time of its correspondence is exactly the time to peak t of P ripple at this acceptance point _p0; Described T _spmaxbe more than or equal to the maximum time difference of P, S ripple in record;

Check the P crest value time: similar with S crest value time check method, with Parabolic Fit time to peak T-X curve, error exceeded to the time to peak value of certain boundary, with the time of matching for time window mid point, reduce search window, again pick up;

Pickup P ripple take-off time in described step (7) also checks that P ripple take-off time realizes like this;

Pickup P ripple take-off time: after the time to peak of P ripple is determined, with time to peak t _p0for in the take-off time detection window of mid point, the long short time-window slided at two respectively calculates the mean value of CFp, STA and LTA, then utilizes the maximal value of STA/LTA as the take-off time t of P ripple _p; Described STA/LTA window can not overlapping (STA and LTA window be chosen like this, suppose that length window length is respectively Ll and Ls, i is the point that will calculate, then the calculation window of LTA is [i-Ll, i], the calculation window of STA be (i, i+Ls],), the ratio of STA/LTA arranges minimum threshold values simultaneously;

Check P ripple take-off time: similar with S ripple take-off time inspection method, with Parabolic Fit take-off time T-X curve, error is exceeded to the take-off time value of certain boundary, with the time of matching for time window mid point, reduce search window, reset the STA/LTA threshold values of a local, again pick up the maximum value of STA/LTA.

Described step (8) is achieved in that

Ineligible P ripple take-off time can be checked out by formula (10):

|(t _s- _tp)-t _s(α-β)／β|≤ε (10)

In formula (10), α and β is the average velocity of P ripple near wave detector and S ripple respectively.The inspection caused in order to release rate error is inaccurate, and a time difference range ε of setting, rejects for the event not meeting formula (10), and this event result of namely having picked up does not retain, and deletes;

Judged by the deviation accumulation of time to peak or the para-curve of take-off time matching and the time to peak of pickup or take-off time, if the value serious offense boundary value of deviation accumulation, then think that this event is false, this event is rejected;

Being calculated as follows of described deviation accumulation:

Suppose M wave detector, the time (time to peak of P ripple or S ripple or take-off time) of pickup is Tpick _j, the time of corresponding matching is Tfit _j, j=1,2,3...M, then cumulative errors

Described boundary value is M*err, err is that picking errors checks boundary.

Described step (10) is achieved in that

If the event detected in step (4), do not pick up in addition, then return step (5) subsequent pick-up, pick up again if do not need, then export and allly meet the S ripple of the condition of step (8) and step (9) and the time to peak of P ripple and take-off time, terminate pick process.

Compared with prior art, the invention has the beneficial effects as follows:

1. the fundamental function of the S ripple improved, has suppressed the energy of P ripple, effectively eliminates the interference of P ripple, enhances the reliability to S ripple event recognition;

2. moveout scan technology makes the energy in-phase stacking of real event part strengthen, and the energy supposition of non-real event is offset, and highlights event energy, adds the accuracy of event recognition.

3. the fundamental function of the P ripple improved, enhance the energy with the P ripple of the S ripple homology of specifying, and the energy of the ripple of other events or seismic phase is pressed, thus effectively can identifies P ripple, and match with the S ripple from same focus;

4. the application of adaptive threshold, further increases the automaticity of method, reduces human factor to the impact of event recognition;

5. special event recognition flow process, effectively identifies the existence of event, the seismic phase of event, when having picked up the walking of event, and the pairing correct to the event from same focus.

6. the method based on energy ratio between the long window and the short window effectively can resist the impact of noise on event take-off time, picks up take-off time more accurately.

7. effective event detection methods, can revise the pickup of irrational event, and can reject the incorrect event of identification.

Accompanying drawing explanation

Fig. 1 is that three-component seismometer places schematic diagram.

Fig. 2 is wave detector horizontal component orientation rotation schematic diagram.

Fig. 3 is X component recording.

Fig. 4 is Y-component record.

Fig. 5 is Z component record.

Fig. 6-1 is the X component recording before direction rotates.

Fig. 6-2 is the postrotational X component recordings in direction.

Fig. 7-1 is the Y-component record before direction rotates.

Fig. 7-2 is the postrotational Y-component records in direction.

Fig. 8-1 is the Z component record before direction rotates.

Fig. 8-2 is the postrotational Z component records in direction.

Fig. 9 is event stack power.

Figure 10 is the micro-seismic event detected by adaptive threshold.

Figure 11-1 is the fundamental function value of S ripple event.

Figure 11-2 is fundamental function values of P ripple event.

Figure 12-1 is the energy ratio between the long window and the short window of S ripple event.

Figure 12-2 is energy ratio between the long window and the short windows of P ripple event.

Figure 13 is a certain acceptance point place record and fundamental function value and energy ratio between the long window and the short window.

Figure 14 is the amplification to Figure 13 boxed area.

Figure 15 is event 1 take-off time pickup result.

Figure 16 is event 2 take-off time pickup result.

Figure 17 is event 3 take-off time pickup result.

Figure 18 is event 4 take-off time pickup result.

Figure 19 is event 5 take-off time pickup result.

Figure 20 is the step block diagram of the inventive method.

Embodiment

Below in conjunction with accompanying drawing, the present invention is described in further detail:

The micro-seismic event P ripple that the present invention utilizes three-component seismometer in well to monitor and S ripple occur in pairs, and S wave energy is stronger, P wave energy is weak, the feature that polarization direction is mutually orthogonal, constructs new S ripple and the fundamental function of P ripple, has invented a kind of micro-seismic event and has detected, seismic phase identification, event is picked up, and the method for event pairing, is parameter when the subsequent treatment such as microearthquake location provide that event is walked accurately.

The identification of the micro-seismic event that three-component seismometer receives in the well that the present invention relates to, carries out describing to the content related in the present invention for three-component seismometer signal here, is easy to promote for other multi-components wave detectors:

(1) calculating of covariance matrix and eigenwert and proper vector

In a sliding window, the number of samples of signal is N, for X, Y, Z tri-components, time window in mathematical expectation be respectively:

$\stackrel{\‾}{x}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{x}_i,\stackrel{\‾}{y}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{y}_i,\stackrel{\‾}{z}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{z}_i;$

Now, covariance matrix (this matrix is existing) just can be write as:

$M=\left[\begin{array}{ccc}\mathrm{\ΣAA}& \mathrm{\ΣAB}& \mathrm{\ΣAC}\\ \mathrm{\ΣBA}& \mathrm{\ΣBB}& \mathrm{\ΣBC}\\ \mathrm{\ΣCA}& \mathrm{\ΣCB}& \mathrm{\ΣCC}\end{array}\right]$

In formula, ∑ is $A=(x_i-\stackrel{\‾}{x}),B=(y_i-\stackrel{\‾}{y}),C=(z_i-\stackrel{\‾}{z}).$

Can in the hope of to three eigenvalue λ by covariance matrix ₁, λ ₂, λ ₃(wherein λ ₁> λ ₂> λ ₃), and correspond to three stack features vectors of eigenwert, wherein eigenvalue of maximum λ ₁characteristic of correspondence vector is V1 (v _x, v _y, v _z).λ ₁the main energy size of wave polarization can be thought, V1 (v _x, v _y, v _z) the main polarization direction of ripple can be thought.Ripple energy distribution in space and particle movement direction just can be represented like this by the eigenwert of covariance matrix and proper vector.

(2) geophone orientation rotates

Fig. 1 is the placement schematic diagram of three-component seismometer in well.Generally, the axis of two horizontal components of three-component seismometer (x ', y ') is random, and in inclined shaft situation, vertical component (z ') is the same with well track, has certain angle.This method must to wave detector spin orientation before event recognition, and the horizontal component in Shi Ge road is axially consistent, and vertical component is vertical.

First introducing ENZ in lower Fig. 1 and 2, Fig. 1 is geographic coordinate system, u, v, w are three coordinate axis of rectangular coordinate system UVW, wherein w and z ' coincidence, uv and x ' y ' is in same level, and u is the intersection of x ' y ' plane and EN plane, and v is the one-component vertical with u in x ' y ' plane.A _x, a _y, a _zthree orthogonal components, wherein a in ENZ coordinate system _xthe projection in EN plane of shooting point position S and acceptance point position R line, a _yin EN plane and a _xvertical component, a _zconsistent with Z axis.θ is the angle of x ' and u, and β is a _xwith the angle of E axle.Fig. 2 is the x ' in Fig. 1, the signal again of y ', u, v place plane.X ', y ' component direction is random and mutually vertical.U, v are also mutually vertical).

Here is the method step of three-component seismometer orientation rotation.

1. composition graphs 1, if three of three-component seismometer record components be respectively x ', y ', z ' (z ' direction is along well track).First under using formula (1) that wave detector choosing is installed to UVW coordinate system:

$\left[\begin{array}{c}u\\ v\\ w\end{array}\right]=\left[\begin{array}{ccc}\cos \mathrm{\θ}& -\sin \mathrm{\θ}& 0\\ \sin \mathrm{\θ}& \cos \mathrm{\θ}& 0\\ 0& 0& 1\end{array}\right]\left[\begin{array}{c}{x}^{\′}\\ y^{\′}\\ z^{\′}\end{array}\right]---\left(1\right)$

2. formula (2) is utilized to continue to rotate u, v, w to ENZ rectangular coordinate system.

$\left[\begin{array}{c}x\\ y\\ z\end{array}\right]=\left[\begin{array}{ccc}{e}_1& e_2& e_3\\ f_1& f_2& f_3\\ g_1& g_2& g_3\end{array}\right]\left[\begin{array}{c}u\\ v\\ w\end{array}\right]---\left(2\right)$

(e in formula ₁, f ₁, g ₁), (e ₂, f ₂, g ₂), (e ₃, f ₃, g ₃) be the direction cosine of UVW tri-axles in ENZ coordinate system respectively.Wherein $(e_3,f_3,g_3)=(\frac{{\mathrm{\Δx}}_1}{d_1},\frac{{\mathrm{\Δy}}_1}{d_1},\frac{{\mathrm{\Δz}}_1}{d_1})(d_1=\sqrt{{\left({\mathrm{\Δx}}_1\right)}^2+{\left({\mathrm{\Δy}}_1\right)}^2+{\left({\mathrm{\Δz}}_1\right)}^2}),$ Δ x ₁, Δ y ₁, Δ z ₁the increment of coordinate of adjacent wave detector in ENZ coordinate system.By UVW orthogonality relation and derive:

$(e_1,f_1,g_1)=(-\frac{f_3}{\sqrt{1-g_3^2}},\frac{e_3}{\sqrt{1-g_3^2}},0)(e_2,f_2,g_2)=(\frac{e_3{g}_3}{\sqrt{1-g_3^2}},\frac{f_3{g}_3}{\sqrt{1-g_3^2}},\sqrt{1-g_3^2})$

3. formula (3) is then utilized to be threaded to by wave detector in the rectangular coordinate system of shooting point and acceptance point line direction formation, if postrotational three vectors are (a _x, a _y, a _z), wherein a _xfor the horizontal direction of line examined by big gun, a _yfor perpendicular to a _xanother transverse axis, a _zvertical is downward.

$\left[\begin{array}{c}{a}_x\\ a_y\\ a_z\end{array}\right]={\left[\begin{array}{ccc}{b}_{11}& b_{12}& 0\\ b_{21}& b_{22}& 0\\ 0& 0& 1\end{array}\right]}^{-1}\left[\begin{array}{c}x\\ y\\ z\end{array}\right]---\left(3\right)$

Make Δ x ₂=x _r-x _s, Δ y ₂=y _r-y _s, Δ z ₂=z _r-z _s; wherein (x _s, y _s, z _s) and (x _r, y _r, z _r) represent shot point and the coordinate of acceptance point in ENZ coordinate system respectively.Then b ₂₁=b ₁₂, b ₂₂=-b ₁₁.

4. formula (4) is utilized to continue rotating vector (a _x, a _y, a _z), make a _xwith due east E to coincidence, a _ywith positive northern N to coincidence, a _xstill downwards vertical;

$\left[\begin{array}{c}X\\ Y\\ Z\end{array}\right]=\left[\begin{array}{ccc}\cos \mathrm{\β}& -\sin \mathrm{\β}& 0\\ \cos \mathrm{\β}& \cos \mathrm{\β}& 0\\ 0& 0& 1\end{array}\right]\left[\begin{array}{c}{a}_x\\ a_y\\ a_z\end{array}\right]---\left(4\right)$

Wherein cos β=Δ x ₂/ d ₂, sin β=Δ y ₂/ d ₂.

5. the calculating of wave detector horizontal azimuth θ can be asked by the method for the component direct P ripple of perforation data being carried out to angle scanning.As postrotational component a _xwith a _ythe maximum angle θ of energy difference be required position angle.That is:

$E\left(\mathrm{\θ}\right)=\underset{k=n1}{\overset{n2}{\mathrm{\Σ}}}{a_x}^2\left(k\right)-\underset{k=n1}{\overset{n2}{\mathrm{\Σ}}}{y_x}^2\left(k\right)---\left(5\right)$

(3) S wave characteristic function

Because micro-seismic event is general all near shooting point, and S wave polarization direction can be thought perpendicular to perforation grid bearing vector S (s _x, s _y, s _z), wherein s _x=x _s-x _,, _y=y _s-y _r, s _z=z _s-z _r, P wave polarization direction is approximately parallel to S (s _x, s _y, s _z), thus utilize with the eigenvalue of maximum λ 1 of covariance matrix and character pair vector V1 (v _x, v _y, v _z), and S (s _x, s _y, s _z) structure S wave characteristic function, such structural attitude function is the interference caused the identification of S ripple to suppress P wave polarization energy.Its constructed fuction is:

$\mathrm{CFs}=(1-|\frac{s_x{v}_x+s_y{v}_y+s_z{v}_z}{\sqrt{{s_x}^2+{s_y}^2+{s_z}^2}+\sqrt{{v_x}^2+{v_y}^2+{v_z}^2}}\left|\right)\mathrm{\λ}1---\left(6\right)$

(4) P wave characteristic function

From P ripple and the S wave polarization direction near normal of same focus, utilize the S wave polarization vector P (p detected _x, p _y, p _z), the eigenvalue of maximum character pair vector of the covariance matrix of the S crest value time point namely picked up; And the eigenvalue of maximum λ 1 of the covariance matrix at some place to be calculated and character pair are to V1 (v _x, v _y, v _z) structure P wave characteristic function, such fundamental function can suppress the energy of the energy of S ripple and the P ripple of other focus, to identify and to pick up the P ripple from same focus.Its constructed fuction is:

$\mathrm{CFp}=(1-|\frac{p_x{v}_x+p_y{v}_y+p_z{v}_z}{\sqrt{{p_x}^2+{p_y}^2+{p_z}^2}+\sqrt{{v_x}^2+{v_y}^2+{v_z}^2}}\left|\right)\mathrm{\λ}1---\left(7\right)$

(5) the moveout scan superposition of event energy

Generally, in well, the energy of three-component S ripple is comparatively strong, and S ripple and P ripple occur all in pairs, and P ripple arrives prior to S ripple, detects the existence of event by detecting S ripple.Concrete steps are as follows:

1. formula (6) is utilized to calculate each acceptance point, the S wave characteristic functional value CFs in each moment.

2. maximum stack energy L (t) in each moment is calculated, wherein

$L\left(t\right)=\underset{j}{\mathrm{\Σ}}{\mathrm{CFs}}_j(t-{\mathrm{\τ}}_j)---\left(8\right)$

τ _jit is the time shift amount of a jth wave detector.This time shift amount on a 3D grid near shooting point, can calculate focus on each grid node by scanning and, to the normal moveout correction time difference of acceptance point, chooses the one group of time difference value making L (t) maximum, as the time shift amount τ of current time _j, and the grid node locations making L (t) maximum is exactly the position of doubtful focus, corresponding moment t _i(i refers to i-th group of event) is exactly a shake time of doubtful focus.

(6) identification of event

After utilizing formula (8) to calculate event stack power value L (t), calculate an adaptive threshold values, if L (t) value in a certain moment is greater than threshold values, then think that at this time, be carved with an event occurs.Concrete steps are as follows:

1. Hilbert transform is carried out to L (t), calculate envelope H (t) of L (t), wherein:

$H\left(t\right)=\sqrt{L^2\left(t\right)+{\hat{L}}^2}\left(t\right)$

Here be the Hilbert transform of L (t), * is convolution symbol.

The calculating of Hilbert transform and its envelope is a kind of known normal signal analytical approach, and formula used is as follows:

$\hat{L}\left(t\right)=L\left(t\right)*\frac{1}{\mathrm{\πt}}$

be the Hilbert transform of L (t), * is convolution symbol.

$H\left(t\right)=\sqrt{L^2\left(t\right)+{\hat{L}}^2\left(t\right)}$

2., in a sliding window, expectation value E (t) and standard variance δ (t) of each moment envelope H (t) is calculated, wherein:

$E\left(t\right)=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{H}_i\left(t\right)$

$\mathrm{\δ}\left(t\right)=\frac{1}{N}\sqrt{\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{(H_i\left(t\right)-E\left(t\right))}^2}$

3. the calculating of adaptive threshold

Ht(t)=E(t-τ)+αδ(t-τ) (9)

Wherein τ is time delay (general being advisable with the half period of S ripple of τ value), and the threshold values caused because of first arrival in order to adjustment changes too early, general being advisable with the one-period of S ripple of τ value.α is the weight coefficient of standard variance, and α is empirical value, and its size is needed to determine by realistic accuracy, specifically determines, generally between 0.5 ~ 2 by recognition effect test.

4. at a regular time length T _event(T _eventlength be greater than in whole microseismograms, the time difference of one group of maximum S ripple and P ripple (from a focus)) in, the maximal value L that L (t) value is greater than Ht (t) will be met _maxt () (" this detection window " just refers to time span T as this detection window _event) in event, corresponding time t _ibe exactly that event plays the shake time, i=1,2,3...K, K are the event numbers detected.In order to avoid P ripple being treated as the S ripple of other events, T _eventbe greater than in whole microseismograms, the time difference of one group of maximum S ripple and P ripple (from a focus).

(7) pickup of S wave-wave group time

1. the pickup of time to peak

Time to peak refers to time corresponding to the ceiling capacity of event.For a certain event that embodiment (6) detects, to each acceptance point, with t-τ _j(t is a shake time of the event that embodiment (6) detects, τ _jthe time shift amount of the jth wave detector calculated in the embodiment (5)) moment is a time to peak searching window mid point, the maximal value corresponding time of searching for CFs at this time in window is exactly the time to peak t of event at this acceptance point _s0.

2. the pickup of take-off time

Take-off time refers to the time that the first arrival ski-jump of event is corresponding.After the time to peak of event is determined, with time to peak t _s0for in the take-off time detection window of mid point, the long short time-window slided at two respectively calculates the mean value of CFs, STA and LTA, then utilizes the maximal value of STA/LTA as the take-off time t of S ripple _s.In order to keep high sensitivity, STA/LTA window can not overlapping (STA and LTA window be chosen like this, suppose long, short time-window length is respectively Ll and Ls, i is the point that will calculate, then the calculation window of LTA is [i-Ll, i], the calculation window of STA is (i, i+Ls]), simultaneously in order to eliminate the pickup of insecure time, the ratio of STA/LTA is provided with minimum threshold values, and (the minimum threshold values of STA/LTA is an empirical value, to be determined by pickup effect test according to the to-noise ratio of data, wherein to-noise ratio is higher, threshold values is less, otherwise threshold values is larger), this threshold values is an empirical value, to be determined by pickup effect test according to the to-noise ratio of data, wherein to-noise ratio is higher, threshold values is less, otherwise threshold values is larger.

(8) pickup of P wave-wave group time

The search of P is the fixed time interval T before the S ripple take-off time determined _spmax(the T in the same embodiment of size (6) _event) in carry out.It is determined by the search P ripple vertical with S wave polarization direction.The fundamental function of this P ripple is at Y _spmaxin interval, with the S ripple polarization vector detected, to each time point, application of formula (7) calculates fundamental function value CFp.

1. the pickup of time to peak

At whole time interval T _spmaxin, the maximal value corresponding time of search CFp is exactly the time to peak t of P ripple at this acceptance point _p0.

2. the pickup of take-off time

After the time to peak of P ripple is determined, with time to peak t _p0for in the take-off time detection window of mid point, the long short time-window slided at two respectively calculates the mean value of CFp, STA and LTA, then utilizes the maximal value of STA/LTA as the take-off time t of P ripple _p.The same with S ripple, in order to keep high sensitivity, STA/LTA window can not be overlapping, and the ratio of STA/LTA arranges minimum threshold values simultaneously, wherein the choosing of STA and LTA window, the same with the pickup of S ripple take-off time with the setting principle of the minimum threshold values of STA/LTA.

(9) marginal testing of event

In order to improve the reliability of event recognition and pickup, check in identification and pick process.

1. the inspection of time to peak

With time to peak matching second-degree parabola equation Tfit of pickup _i=ax _i ²+ bx _i+ c, (x _ifor receiving period, Tfit _itime for matching), calculate the match value of each acceptance point and the error e rr=|Tfit-Tpick| (Tpick is actual pickup value) of actual pickup value.When error e rr exceedes certain boundary value, with the time of matching for time window mid point, reduce search window, the time to peak of the maximum value of again picking up fundamental function value this acceptance point the most.Wherein the boundary value of err is generally several sampled point time, the error precision of this value required for reality, is determined by pickup effect test.

2. the inspection of take-off time

The same with the process of matching time to peak, the parabolic equation of a matching take-off time, equally, for take-off time value error being exceeded to certain boundary, with the time of matching for time window mid point, reduce search window, reset (when the setting principle of this threshold values picks up with S ripple take-off time, STA/LTA threshold values principle being set the same) of a local, again pick up the maximum value of the STA/LTA in wicket.

3. P, S ripple matching check

Ineligible P ripple take-off time can be checked out by formula (10).

|(t _s-t _p)-t _s(α-β)／β|≤ε (10)

α and β is known, is the average velocity (P ripple and S ripple average velocity are in advance known quantity) of P ripple near wave detector and S ripple respectively.The inspection caused in order to release rate error is inaccurate, a less time difference range ε arranging (" time difference range ε " carries out adjusting determining according to the precision of speed and pickup effect), rejects for the event of the condition not meeting formula (10).

4. the true and false inspection of event

Check at event time to peak and in take-off time checking process, first can check the true and false of event.This is judged by the deviation accumulation of the para-curve of time to peak (or take-off time) matching and the time to peak (or take-off time) of pickup, if deviation accumulation is bigger than a certain value, think that this event is incorrect, then pickup result is disallowable.

Wherein cumulative errors pass through formula calculate, j=1,2,3...M, M are wave detector numbers, Tpick _jthe time (time to peak of P ripple or S ripple or take-off time) of pickup, Tfit _jit is corresponding fit time.Deviation accumulation value boundary is generally M*err, err is that step peak error 1. checks boundary.

(10) event recognition and the overall realization flow of pickup, as shown in figure 20, comprising:

1. step each acceptance point to event monitoring record of application implementation mode (2) carries out orientation rotation, makes X component and due east E to coincidence, and Y-component and positive northern N are to coincidence, and Z component is vertical.

2. to postrotational record, the fundamental function formula (6) of embodiment (3) is utilized to calculate each acceptance point, the fundamental function value CFs in each moment.

3. event stack power function L (t) is calculated according to the step of embodiment (5).

4. a shake time t of all Possible events is detected according to the step of embodiment (6) _i.

5. 1. pick up each acceptance point time to peak of one group of S ripple according to the step of embodiment (7), then 1. check the S crest value time according to the step of embodiment (9); 2. pick up S ripple take-off time according to the step of embodiment (7), then 2. check the S crest value time according to the step of embodiment (9).

6. the fixed time interval T before the one group of S ripple take-off time 5. determined in flow process _spmaxinterior each time point for each acceptance point, application implementation scheme (4) fundamental function formula (7) calculates fundamental function value CFp.

7. 1. pick up each acceptance point time to peak of one group of P ripple according to the step of embodiment (8), then 1. check the P crest value time according to the step of embodiment (9); 2. pick up P ripple take-off time according to the step of embodiment (8), then 2. check P ripple take-off time according to the step of embodiment (9).

8. 3. check one group of P, S ripple match condition of having picked up according to the step of embodiment (9), if ineligible, then reject this event.

9. 4. check the true and false situation of one group of event of having picked up according to the step of embodiment (9), if ineligible, then reject this event.

10. Returning process 3., and next group event of subsequent pick-up, until all events have been picked up, terminates pickup.

One embodiment of the present of invention are as follows:

Defeat with a bite WIH and split micro-seismic monitoring data for example.16 grades of three-components receive, and inspection well is straight well, well head relative coordinate: 0,0,0; First wave detector degree of depth 790m, phone spacing 10m; Shooting point position: 292.68,288.82,1243.18;

Accompanying drawing 3,4,5 is through the X of a certain section of monitoring record of denoising respectively, Y, Z tri-components.

Carry out the rotation of geophone orientation angle to these three components, accompanying drawing 6-1 to Fig. 8-2 is through the X that geophone orientation rotates front and back respectively, and the comparison diagram of Y, Z tri-component recordings, left side is the record before orientation rotation, and right side is the record after orientation rotation.Can see from postrotational effect, after horizontal direction rotates, X, Y, under the polarization direction, each road of Z component has been corrected to the same coordinate system system.

Fig. 9 carries out moveout scan to postrotational monitoring record to superpose the event monitoring stack power obtained.5 obvious micro-seismic event have been can be clearly seen that from stack power figure.

Calculate adaptive threshold values to stack power, automatically detect micro-seismic event, Figure 10 is moveout scan stack power, energy envelope and adaptive threshold curve, and the event relation figure monitored.Wherein solid line is adaptive threshold, and long dotted line is energy envelope, and short dash line is stack power, and black side's point is the doubtful event of being monitored out by adaptive threshold.See from figure, in the event (black silk Fang Dian represents event) be detected except 5 significantly except, also has some other doubtful event, to these doubtful events owing to not being real event, the result of carrying out picking up is random, do not have rule to follow, in the event inspection therefore below can disallowable fall.

Shown in Figure 11-1, Figure 11-2 and 12-1, Figure 12-2 is long fundamental function value and the energy ratio between the long window and the short window value of wherein one group of S ripple from same event and the P ripple detected, window 40ms, short time-window 15ms time wherein long; Region folded by dotted line is time to peak and the take-off time value search window of event respectively, see from figure, the energy of S ripple and P ripple by long short time-window ratio afterwards, peak value is more precipitous, as long as choose the maximal value corresponding time that an appropriate window (region in figure folded by dotted line) searches out energy Ratios, just obtain the take-off time of ripple.

Figure 13 is three component recording X of a certain acceptance point intercepted, Y, Z, three trace records giving first acceptance point place are corresponding, S wave characteristic functional value (CFS), P wave characteristic functional value (CFP), S wavelength short time-window energy ratio (STA/LTA (S)), P wave-wave energy ratio between the long window and the short window value (STA/LTA (P)), Figure 14 amplifies last event (in accompanying drawing 13 square frame event) wherein, indicated the search window scope of S ripple and P ripple respectively with arrow in figure, and S ripple and P crest value time are in the take-off location of take-off time, the length of LTA and STA indication line segment denotes length (window 40ms time long of long short time-window respectively, short time-window 15ms).From the figure after amplification, see that P ripple arrives prior to S ripple, the crest of the corresponding S ripple of maximal value of CFS, the corresponding P wave-wave peak of maximal value of CFP, the maximal value corresponding S ripple take-off time of STA/LTA (S), the maximal value corresponding P ripple take-off time of STA/LTA (P), wherein the variation tendency of STA/LTA value is more stable, only has the spike that is single, and the change of fundamental function value is along with the continuity of period of wave, around main peak, have the other peak that some are little.The searching window of S ripple is a bit larger tham its cycle, and the searching window of P ripple is a bit larger tham forward the time difference of the P ripple of S ripple from the take-off time of S ripple.

Figure 15 to Figure 19 be five groups of obvious events take-off time pickup result (every three roads are an acceptance point, and three roads are respectively X from top to bottom, Y, Z component), wherein in same event, what first arrive is P ripple, rear arrival be S ripple.From the result of pickup, this method can identify the existence of event accurately, judges the seismic phase of S ripple from same focus and P ripple, and when event of picking up accurately is walked.

The energy relationship of the S ripple that the present invention occurs in pairs according to micro-seismic event and P ripple and polarization characteristic, improve the fundamental function of S and P ripple, the corresponding identification of the event that also improves and the method for pickup and flow process.Comprise: identify S ripple event by the moveout scan superposition of the fundamental function value of S ripple and adaptive threshold; P ripple event is identified by the fundamental function value of the P ripple with S wave polarization Attribute Association; Time to peak and the take-off time of event is picked up respectively by the extreme value of fundamental function value in a timing window and the extreme value of energy ratio between the long window and the short window; By effective event detection methods, ensure the reliability of event.This invention effectively can identify the seismic phase of micro-seismic event, matches, and pick up micro-seismic event to the event of homology, is parameter when micro-seismic monitoring subsequent treatment provides event to walk.

Technique scheme is one embodiment of the present invention, for those skilled in the art, on the basis that the invention discloses application process and principle, be easy to make various types of improvement or distortion, and the method be not limited only to described by the above-mentioned embodiment of the present invention, therefore previously described mode is just preferred, and does not have restrictive meaning.

Claims (10)

1. a method for borehole microseismic event recognition and pickup, is characterized in that: said method comprising the steps of:

(1) carry out orientation rotation to each acceptance point of event monitoring record, make X component and due east E to coincidence, Y-component and positive northern N are to coincidence, and Z component is vertical;

(2) to postrotational record, the fundamental function value CFs in each each moment of acceptance point is calculated;

(3) by moveout scan technique computes event stack power function L (t);

(4) a shake time t of event is automatically detected by adaptive threshold _i:

(5) pick up the S crest value time of one group of event and check the S crest value time; Pickup S ripple take-off time also checks S ripple take-off time;

(6) for each acceptance point, the fixed time interval T before the one group of S ripple take-off time determined _spmaxinterior to each time point calculating fundamental function value CFp;

(7) pick up each acceptance point time to peak of the P ripple of one group of event and check the P crest value time; Pickup P ripple take-off time also checks P ripple take-off time;

(8) check P ripple, the S ripple match condition of one group of event of having picked up, if ineligible, then reject this event; Check the true and false situation of one group of event of having picked up, if ineligible, then reject this event;

(9) judge whether that all events have all been picked up, if not, then return step (5), if so, then export peak time and the take-off time of all qualified S ripples and P ripple, terminate pick process.

2. the method for borehole microseismic event recognition according to claim 1 and pickup, is characterized in that: described step (1) comprising:

(11) establish three of three-component seismometer record components be respectively x ', y ', z ', first use formula (1) by wave detector choosing install to UVW coordinate system under:

$\left[\begin{array}{c}u\\ v\\ w\end{array}\right]=\left[\begin{array}{ccc}\cos \mathrm{\θ}& -\sin \mathrm{\θ}& 0\\ \sin \mathrm{\θ}& \cos \mathrm{\θ}& 0\\ 0& 0& 1\end{array}\right]\left[\begin{array}{c}{x}^{\′}\\ y^{\′}\\ z^{\′}\end{array}\right]---\left(1\right)$

Here θ represents the angle of u and x ';

(12) formula (2) is utilized to continue to rotate u, v, w to ENZ rectangular coordinate system:

$\left[\begin{array}{c}x\\ y\\ z\end{array}\right]=\left[\begin{array}{ccc}{e}_1& e_2& e_3\\ f_1& f_2& f_3\\ g_1& g_2& g_3\end{array}\right]\left[\begin{array}{c}u\\ v\\ w\end{array}\right]---\left(2\right)$

(e in formula ₁, f ₁, g ₁), (e ₂, f ₂, g ₂), (e ₃, f ₃, g ₃) be the direction cosine of UVW tri-axles in ENZ coordinate system respectively;

Wherein $(e_3,f_3,g_3)=(\frac{{\mathrm{\Δx}}_1}{d_1},\frac{{\mathrm{\Δy}}_1}{d_1},\frac{{\mathrm{\Δz}}_1}{d_1})(d_1=\sqrt{{\left({\mathrm{\Δx}}_1\right)}^2+{\left({\mathrm{\Δy}}_1\right)}^2+{\left({\mathrm{\Δz}}_1\right)}^2}),$ Δ x ₁, Δ y ₁, Δ z ₁the increment of coordinate of adjacent wave detector in ENZ coordinate system, by UVW orthogonality relation and derive:

$(e_1,f_1,g_1)=(-\frac{f_3}{\sqrt{1-g_3^2}},\frac{e_3}{\sqrt{1-g_3^2}},0)(e_2,f_2,g_2)=(\frac{e_3{g}_3}{\sqrt{1-g_3^2}},\frac{f_3{g}_3}{\sqrt{1-g_3^2}},\sqrt{1-g_3^2})$

(13) formula (3) is utilized to be threaded to by wave detector in the rectangular coordinate system of shooting point and acceptance point line direction formation, if postrotational three vectors are (a _x, a _y, a _z), wherein a _xfor the horizontal direction of line examined by big gun, a _yfor perpendicular to a _xanother transverse axis, a _zvertical is downward:

$\left[\begin{array}{c}{a}_x\\ a_y\\ a_z\end{array}\right]={\left[\begin{array}{ccc}{b}_{11}& b_{12}& 0\\ b_{21}& b_{22}& 0\\ 0& 0& 1\end{array}\right]}^{-1}\left[\begin{array}{c}x\\ y\\ z\end{array}\right]---\left(3\right)$

Make Δ x ₂=x _r-x _s, Δ y ₂=y _r-y _s, Δ z ₂=z _r-z _s; wherein (x _s, y _s, z _s) and (x _r, y _r, z _r) represent shooting point and the coordinate of acceptance point in ENZ coordinate system respectively; Then b ₂₁=b ₁₂, b ₂₂=-b ₁₁;

(14) formula (4) is utilized to continue rotating vector (a _x, a _y, a _z), make a _xwith due east E to coincidence, a _ywith positive northern N to coincidence, a _zstill downwards vertical;

$\left[\begin{array}{c}X\\ Y\\ Z\end{array}\right]=\left[\begin{array}{ccc}\cos \mathrm{\β}& -\sin \mathrm{\β}& 0\\ \cos \mathrm{\β}& \cos \mathrm{\β}& 0\\ 0& 0& 1\end{array}\right]\left[\begin{array}{c}{a}_x\\ a_y\\ a_z\end{array}\right]---\left(4\right)$

Wherein β is a _xwith the angle of E, cos β=Δ x ₂/ d ₂, sin β=Δ y ₂/ d ₂;

(15) method by carrying out angle scanning to the component direct P ripple of perforation data tries to achieve wave detector horizontal azimuth: as postrotational component a _xwith a _ythe maximum angle θ of energy difference be required position angle:

$E\left(\mathrm{\θ}\right)=\underset{k=n1}{\overset{n2}{\mathrm{\Σ}}}{a_x}^2\left(k\right)-\underset{k=n1}{\overset{n2}{\mathrm{\Σ}}}{y_x}^2\left(k\right)$

3. the method for borehole microseismic event recognition according to claim 2 and pickup, is characterized in that: described step (2) is achieved in that

(21) in a sliding window, for these three components of postrotational X, Y, Z, its covariance matrix is built:

$M=\left[\begin{array}{ccc}\mathrm{\ΣAA}& \mathrm{\ΣAB}& \mathrm{\ΣAC}\\ \mathrm{\ΣBA}& \mathrm{\ΣBB}& \mathrm{\ΣBC}\\ \mathrm{\ΣCA}& \mathrm{\ΣCB}& \mathrm{\ΣCC}\end{array}\right]$

In formula, ∑ is $A=(x_i-\stackrel{\‾}{x}),B=(y_i-\stackrel{\‾}{y}),C=(z_i-\stackrel{\‾}{z}),$ Wherein $\stackrel{\‾}{x}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{x}_i,$

$\stackrel{\‾}{y}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{y}_i,\stackrel{\‾}{z}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{z}_i;$ Asked by covariance matrix and obtain three eigenvalue λ ₁, λ ₂, λ ₃and correspond to three stack features vectors of eigenwert, wherein λ ₁> λ ₂> λ ₃, eigenvalue of maximum λ ₁characteristic of correspondence vector is V1 (v _x, v _y, v _z);

(22) utilize with the eigenvalue of maximum λ 1 of covariance matrix and its characteristic of correspondence vector V1 (v _x, v _y, v _z), and perforation grid bearing vector S (s _x, s _y, s _z) structure S wave characteristic function:

$\mathrm{CFs}=(1-|\frac{s_x{v}_x+s_y{v}_y+s_z{v}_z}{\sqrt{{s_x}^2+{s_y}^2+{s_z}^2}+\sqrt{{v_x}^2+{v_y}^2+{v_z}^2}}\left|\right)\mathrm{\λ}1---\left(6\right)$

Wherein: S (s _x, s _y, s _z)=(x _s-x _r, y _s-y _r, z _s-z _r)

Then formula (6) is utilized to calculate the fundamental function value CFs in each each moment of acceptance point to postrotational record (X, Y, Z).

4. the method for borehole microseismic event recognition according to claim 3 and pickup, is characterized in that: described step (3) is achieved in that

Formula (8) is utilized to calculate maximum stack energy L (t) in each moment:

$L\left(t\right)=\underset{j}{\mathrm{\Σ}}{\mathrm{CFs}}_j(t-{\mathrm{\τ}}_j)---\left(8\right)$

In formula (8), τ _jit is the time shift amount of a jth wave detector.

5. the method for borehole microseismic event recognition according to claim 4 and pickup, is characterized in that: described step (4) comprising:

(41) Hilbert transform is carried out to L (t), calculate envelope H (t) of L (t);

(42) in a sliding window, expectation value E (t) and standard variance δ (t) of each moment envelope H (t) is calculated;

(43) formula (9) is utilized to carry out the calculating of adaptive threshold:

Ht(t)=E(t-τ)+αδ(t-τ) (9)

In formula (9), τ is time delay, and the threshold values caused because of first arrival in order to adjustment changes too early.α is the weight coefficient of standard variance;

(44) at a regular time length T _eventin, the maximal value L that L (t) value is greater than Ht (t) will be met _max(t) as the event in this detection window, corresponding time t _ibe exactly that event plays the shake time, i=1,2,3...K, K are the event numbers detected, described T _eventbe greater than the maximum time difference of S-P ripple; The length of described detection window is time span T _event.

6. the method for borehole microseismic event recognition according to claim 5 and pickup, is characterized in that: the pickup S crest value time in described step (5) also checks that the S crest value time is achieved in that

The pickup S crest value time: the S crest value time refers to time corresponding to the ceiling capacity of event, to a certain event that step (4) detects, to each acceptance point, with t _i-τ _jmoment is this acceptance point time to peak searching window mid point, searches for the maximal value of CFs at this time in window, and its corresponding time is exactly the time to peak t of event at this acceptance point _s0; t _ia shake time of i-th group of event that step (4) detects, τ _jit is the time shift amount of the jth wave detector calculated in step (3);

Check the S crest value time: with parabolic equation matching time to peak T-X curve, for the time to peak value that error is beyond the mark, with the theoretical peak time Tfit with Parabolic Fit _ifor time window mid point, reduce search window, then in the search window reset, find out the maximal value of CFs;

Pickup S ripple take-off time in described step (5) also checks that S ripple take-off time is achieved in that

Pickup S ripple take-off time: S ripple take-off time refers to the time that the first arrival ski-jump of event is corresponding; After the time to peak of event is determined, with time to peak t _s0for in the take-off time detection window of mid point, the long short time-window slided at two respectively calculates mean value STA and LTA of CFs, then utilizes the maximal value of STA/LTA as the take-off time t of S ripple _s; STA/LTA window can not be overlapping, simultaneously for the ratio of STA/LTA is provided with minimum threshold values;

Check S ripple take-off time: with Parabolic Fit take-off time T-X curve, for the take-off time value that error is beyond the mark, with the theoretical take-off time Tfit with Parabolic Fit _ifor time window mid point, reduce search window, reset the STA/LTA threshold values of a local, in the search window reset, again find out the maximal value of STA/LTA.

7. the method for borehole microseismic event recognition according to claim 6 and pickup, is characterized in that: described step (6) is achieved in that

For an acceptance point, the fixed time interval T before the S ripple take-off time determined _spmaxin, utilize the S wave polarization vector P (p detected _x, p _y, p _z), and the eigenvalue of maximum λ 1 of the covariance matrix at time point place to be calculated and character pair vector V1 (v _x, v _y, v _z) structure P wave characteristic function:

$\mathrm{CFp}=(1-|\frac{p_x{v}_x+p_y{v}_y+p_z{v}_z}{\sqrt{{p_x}^2+{p_y}^2+{p_z}^2}+\sqrt{{v_x}^2+{v_y}^2+{v_z}^2}}\left|\right)\mathrm{\λ}1---\left(7\right)$

Then formula (7) is utilized to calculate CFp to each time point of each acceptance point.

8. the method for borehole microseismic event recognition according to claim 7 and pickup, is characterized in that: the pickup P crest value time in described step (7) also checks that the P crest value time is achieved in that

The pickup P crest value time: at whole time interval T _spmaxin, the maximal value of search CFp, the time of its correspondence is exactly the time to peak t of P ripple at this acceptance point _p0; Described T _spmaxbe more than or equal to the maximum time difference of P, S ripple in record;

Check the P crest value time: with Parabolic Fit time to peak T-X curve, error is exceeded to the time to peak value of certain boundary, with the time of matching for time window mid point, reduce search window, again pick up.

Pickup P ripple take-off time in described step (7) also checks that P ripple take-off time realizes like this;

Pickup P ripple take-off time: after the time to peak of P ripple is determined, with time to peak t _p0for in the take-off time detection window of mid point, the long short time-window slided at two respectively calculates the mean value of CFp, STA and LTA, then utilizes the maximal value of STA/LTA as the take-off time t of P ripple _p; Described STA/LTA window can not overlapping (STA and LTA window be chosen like this, suppose that length window length is respectively Ll and Ls, i is the point that will calculate, then the calculation window of LTA is [i-Ll, i], the calculation window of STA be (i, i+Ls],), the ratio of STA/LTA arranges minimum threshold values simultaneously;

Check P ripple take-off time: similar with S ripple take-off time inspection method, with Parabolic Fit take-off time T-X curve, error is exceeded to the take-off time value of certain boundary, with the time of matching for time window mid point, reduce search window, reset the STA/LTA threshold values of a local, again pick up the maximum value of STA/LTA.

9. the method for borehole microseismic event recognition according to claim 8 and pickup, is characterized in that: described step (8) is achieved in that

Ineligible P ripple take-off time can be checked out by formula (10):

|(t _s-t _p)-t _s(α-β)／β|≤ε (10)

In formula (10), α and β is the average velocity of P ripple near wave detector and S ripple respectively, arranges a time difference range ε, rejects for the event not meeting formula (10), this event result of namely having picked up does not retain, and deletes;

Judged by the deviation accumulation of time to peak or the para-curve of take-off time matching and the time to peak of pickup or take-off time, if the value serious offense boundary value of deviation accumulation, then think that this event is false, this event is rejected;

Being calculated as follows of described deviation accumulation:

Suppose M wave detector, the time of pickup is Tpick _j, the time of corresponding matching is Tfit _j, j=1,2,3...M, then cumulative errors

Described boundary value is M*err, err is that picking errors checks boundary.

10. the method for borehole microseismic event recognition according to claim 1 and pickup, is characterized in that: described step (9) is achieved in that

If the event detected in step (4), do not pick up in addition, then return step (5) subsequent pick-up, pick up again if do not need, then export and allly meet the S ripple of the condition of step (8) and step (9) and the time to peak of P ripple and take-off time, terminate pick process.