CN104516015B - A kind of method for determining coal bed gas compressional wave and shear wave velocity - Google Patents

Abstract

The invention belongs to exploration data processing procedure, is a kind of method for determining coal bed gas compressional wave and shear wave velocity.Shear wave velocity and velocity of longitudinal wave are obtained after processing seismic prospecting and log data, respectively velocity of longitudinal wave and shear wave velocity are processed, meansigma methodss are made to be respectively zero, group compressional wave and shear wave velocity are normalized, ask for compressional wave and shear wave velocity optimal mapping, iteration obtains optimal mapping functional value, returns and obtains functional relation, according to shear wave or velocity of longitudinal wave that functional relation obtains other wells or area.Instant invention overcomes traditional multiple regression procedure needs the defect for presupposing a functional relationship, the correlation coefficient of coal bed gas shear wave and velocity of longitudinal wave after returning, the shear wave or velocity of longitudinal wave that another mouthful of well or area can be calculated using the compressional wave and shear wave velocity relation that obtain can be effectively improved.

Description

A kind of method for determining coal bed gas compressional wave and shear wave velocity

Technical field

The present invention relates to geophysical exploration technology, belongs to exploration data processing procedure, it is that one kind utilizes optimal mapping side The method that method determines coal bed gas compressional wave and shear wave velocity.

Background technology

With the development of energy industry, the exploration and development of coal bed gas is increasingly paid attention to, and this is just physical prospecting and logging technique Propose new challenge.Because there is huge difference with conventional gas at the aspect such as generation and storage in coal bed gas so that be directed to Many formula and research method of natural gas pool are no longer suitable in terms of the description and research of bed gas reservoir, and this just promotes people to adopt Coal bed gas is studied with new method.

Research for bed gas reservoir gradually increased in recent years, when the empirical equation for being suitable for coal bed gas is found, Ren Menchang Relation between the physical parameter obtained come statistical analysiss seismic prospecting and well logging using multiple regression procedure.Traditional polynary time Return method to need first to assume a functional relationship, and for the compressional wave and shear wave to propagating in coal bed gas, its speed is subject to rock The impact of physical property and gas saturation etc., relation are extremely complex, it is difficult to previously given this functional relationship, this frequently results in parameter Return failure, it is impossible to accurately determine the relation of the compressional wave and shear wave velocity of coal bed gas, so as to be difficult to determine compressional wave and shear wave velocity.

Content of the invention

It is an object of the present invention to provide a kind of can obtain compressional wave and the relation after shear wave velocity optimal mapping, can be indulged The relation of ripple and shear wave velocity and respective optimal mapping, while other wells or regional compressional wave or shear wave can be calculated according to the relation The method that the utilization optimal mapping method of speed determines coal bed gas compressional wave and shear wave velocity.

The present invention is realized by following steps：

1）Collection seismic prospecting and log data, obtain shear wave velocity y and velocity of longitudinal wave x after process;

2）Their meansigma methodss are made to be respectively zero velocity of longitudinal wave x and shear wave velocity y process respectively according to the following formula：

In formula：L is a group number, and j is the group serial number of compressional wave and shear wave velocity x and y, x_jAnd y_jIt is compressional wave and shear wave respectively Speed x and j-th element of y

3）According to the following formula L groups compressional wave and shear wave velocity are normalized：

In formula, L is a group number；J is the group serial number of compressional wave and shear wave velocity x and y；x_jAnd y_jIt is compressional wave and shear wave respectively Speed x and j-th element of y;Max (x) is the maximum for asking for velocity of longitudinal wave x；Max (y) is the maximum for asking for shear wave velocity y Value；

4）The initial value φ of compressional wave and shear wave velocity x and y optimal mapping is calculated as the following formula respectively⁰(x) and θ⁰(y)：

In formula, L is the group number of compressional wave and shear wave velocity x and y；J is the group serial number of compressional wave and shear wave velocity x and y；x_jWith y_jIt is j-th element of compressional wave and shear wave velocity x and y respectively.

5）The optimal mapping φ of compressional wave and shear wave velocity x and y is asked for according to the following formula^k(x) and θ^k(y)：

In formula, k is iterationses；S refers to and data is smoothed that smoothing method is using splines method or super smooth side Method；x_jAnd y_jIt is j-th element of compressional wave and shear wave velocity x and y respectively；φ^k(x_j) and θ^k(y_j) it is kth time iteration respectively, Compressional wave and shear wave velocity optimal mapping φ^k(x) and θ^kJ-th element of (y).

6）It is calculated as follows iteration error：

In formula：K is iterationses；

7）Repeat step 5）With 6）, until adjacent iteration error twice meets following formula：

|e²[θ^k(y_j),φ^k(x_j)]-e²[θ^k-1(y_j),φ^k-1(x_j)]|≤δ （6）

In formula, k is iterationses；δ is error threshold value, and span is 0.001 to 0.01, most preferably 0.001；Work as phase Adjacent iteration error twice meets formula（6）When, kth time iteration obtains the optimal mapping function that θ (y) is referred to as shear wave velocity y, obtains φ (x) be referred to as velocity of longitudinal wave x optimal mapping function；φ^k(x_j) and θ^k(y_j) it is kth time iteration, compressional wave and shear wave speed respectively Degree optimal mapping φ^k(x) and θ^kJ-th element of (y).

8）By step 7）Optimal mapping function # (y) for obtaining and the functional value of φ (x), are carried out using linear regression method Return, obtain the functional relation of θ (y) and φ (x)：

θ(y)=f(φ(x)) （7）

In formula, f be by linear regression obtain method to θ (y) and φ (x) functional relation, y and x be respectively shear wave speed Degree y and the set of velocity of longitudinal wave；

9）According to functional relation θ (y), φ (x) and f, the shear wave or velocity of longitudinal wave in other wells or area is calculated.

Step 9）Calculating process be：First other wells or the velocity of longitudinal wave in area（Or shear wave velocity）Substitute into function to close It is formula φ (x) (or θ (y)）, obtain optimal mapping (or the optimal mapping of shear wave velocity of velocity of longitudinal wave）, further according to formula（7） Ask for optimal mapping (or the optimal mapping of velocity of longitudinal wave of shear wave velocity）, finally according to functional relation θ (y) reverse shear wave speed Degree (or according to functional relation φ (x) reverse velocity of longitudinal wave）.

The present invention overcomes biography by optimal mapping homing method statistical analysiss coal bed gas shear wave and the relation of velocity of longitudinal wave System multiple regression procedure needs the defect for presupposing a functional relationship.After this method can effectively improve recurrence, coal bed gas is horizontal Ripple and the correlation coefficient of velocity of longitudinal wave, at the same can using the compressional wave that obtains and shear wave velocity relation, calculate another mouthful of well or The shear wave or velocity of longitudinal wave in area.

Description of the drawings

Fig. 1 is the scatterplot of the shear wave velocity and velocity of longitudinal wave of experiment well of the present invention（Straight line represents the pass of Linear Quasi sum System）；

Fig. 2 is the scatterplot of experiment well shear wave velocity optimal mapping of the present invention and shear wave velocity；

Fig. 3 is the scatterplot of experiment well velocity of longitudinal wave optimal mapping of the present invention and velocity of longitudinal wave；

Fig. 4 is the scatterplot of experiment well shear wave velocity optimal mapping of the present invention and velocity of longitudinal wave optimal mapping（Straight line is represented The relation of Linear Quasi sum）；

Fig. 5 is the shear wave velocity that the relation obtained using experiment well of the present invention calculates another well of present invention experiment（Thick line）With The shear wave velocity of actual measurement（Fine rule）Contrast；

Fig. 6 is the shear wave of the shear wave velocity and another well actual measurement that the relation obtained using experiment well of the present invention calculates another well The percentage error of speed.

Specific embodiment

The present invention is first processed to all of geological data and log data, obtains shear wave velocity and velocity of longitudinal wave, right These data carry out meansigma methodss for zero-sum normalized, expect that method asks for the optimal of different physical parameters using give-and-take conditions Conversion, is carried out returning the relation for obtaining the two to the optimal mapping of shear wave and velocity of longitudinal wave using linear regression method, last profit The shear wave or velocity of longitudinal wave of other wells or area are calculated with this relation for obtaining.

The specific embodiment of the present invention is as follows：

1）Collection seismic prospecting and log data, obtain L group shear wave velocity y and shear wave velocity x after process（As follows）;

2）According to formula（1）It is zero process to carry out meansigma methodss to y and x；

3）According to formula（2）Y and x are normalized；

4）According to formula（3）The initial value φ of x and y optimal mapping is calculated respectively⁰(x) and θ⁰(y)：

5）The optimal mapping φ of the x and y of kth time iteration is asked for according to formula (4)^k(x) and θ^k(y);

6）According to formula（5）The error of adjacent iteration twice is calculated,

7）Repeat step 5）With 6）, until adjacent iteration error twice meets formula（6）, at this moment obtain shear wave velocity y's Optimal mapping function # (y), and optimal mapping function phi (x) of velocity of longitudinal wave x

8）Optimal mapping function # (y) and the functional value of φ (x), returned using linear regression method, obtained θ (y) Functional relation f with φ (x)；

9）According to functional relation θ (y), φ (x) and f, the shear wave or velocity of longitudinal wave in other wells or area is calculated： First other wells or the velocity of longitudinal wave in area（Or shear wave velocity）Substitution functional relation φ (x) (or θ (y)）, obtain compressional wave The optimal mapping (or optimal mapping of shear wave velocity) of speed, further according to formula（7）The optimal mapping for asking for shear wave velocity is (or vertical The optimal mapping of wave velocity）, finally according to functional relation θ (y) reverse shear wave velocity (or according to functional relation φ (x) Reverse velocity of longitudinal wave）.

The following is instantiation of the present invention：

1) prestack seismogram and well-logging is obtained, after process, obtains shear wave velocity y and shear wave velocity x；

Fig. 1 is the scatterplot of the shear wave velocity and velocity of longitudinal wave of experiment well of the present invention, present invention experiment embodiment the most.In figure Straight line represents the relation of Linear Quasi sum, it can be seen that correlation coefficient is 0.889.

2) according to step（2）Step（3）Pretreatment is carried out to y and x, meansigma methodss is obtained for the shear wave after zero-sum normalization And velocity of longitudinal wave；

3）According to step（4）Step（7）Try to achieve the optimal mapping of y and x；

Fig. 2 is the scatterplot of experiment well shear wave velocity optimal mapping of the present invention and shear wave velocity, and Fig. 3 is experiment well of the present invention Velocity of longitudinal wave optimal mapping and the scatterplot of velocity of longitudinal wave, it can be seen that either shear wave velocity or the optimal change of velocity of longitudinal wave Change all more complicated with the relation of its own, it is difficult to Linear Quasi and representing.

4）According to step（8）Respectively shear wave velocity optimal mapping θ (y) and velocity of longitudinal wave optimal mapping φ (x), using line Property homing method is returned, and obtains the functional relation f of θ (y) and φ (x)

Fig. 4 is experiment well shear wave velocity optimal mapping of the present invention and the scatterplot after velocity of longitudinal wave optimal mapping, straight in figure Line represents the relation of Linear Quasi sum（That is functional relation f）, it can be seen that after horizontal, compressional wave optimal mapping, correlation coefficient increases Plus, reach 0.974.

5）According to step 6 and 7）The compressional wave for obtaining and shear wave velocity optimal mapping function # (y) and φ (x), and step 8） Functional relationship f for obtaining, calculates the shear wave or velocity of longitudinal wave in another mouthful of well or another area.

Fig. 5 is the relation according to experiment well shear wave velocity optimal mapping of the present invention and velocity of longitudinal wave optimal mapping, using this Invent the shear wave velocity that another experiment well velocity o P wave is obtained（Thick line）, and the shear wave velocity of experiment well of the present invention actual measurement （Fine rule）Contrast.Fig. 6 is that the relation obtained using present invention experiment calculates another experiment well shear wave velocity of the invention and actual measurement The percentage error of shear wave velocity.As can be seen that utilizing the horizontal stroke of another well of the relation calculating of experiment well of the present invention from contrast The wave velocity shear wave velocity actual with the well is basically identical, and calculated shear wave velocity is exhausted with the percentage error of actual speed Most of needs for being less than 5%, subsequent treatment and explanation being met.

Claims (2)

1. a kind of method for determining coal bed gas compressional wave and shear wave velocity, feature is through the following steps that realize：

1) collection seismic prospecting and log data, obtain velocity of longitudinal wave x and shear wave velocity y after process；

2) their meansigma methodss are made to be respectively zero respectively to velocity of longitudinal wave x and shear wave velocity y process according to the following formula：

$\left\{\begin{array}{c}{x}_j=x_j-\left(\underset{j=1}{\overset{L}{\mathrm{\Σ}}}{x}_j\right)/L\\ y_j=y_j-\left(\underset{j=1}{\overset{L}{\mathrm{\Σ}}}{y}_j\right)/L\end{array}\right.---\left(1\right)$

In formula：L is a group number, and j is the group serial number of velocity of longitudinal wave x and shear wave velocity y, and xj and yj is velocity of longitudinal wave x and horizontal stroke respectively J-th element of wave velocity y；

3) according to the following formula L group velocity of longitudinal wave x and shear wave velocity y are normalized：

$\left\{\begin{array}{c}{x}_j=x_j/\max \left(x\right)\\ y_j=y_j/\max \left(y\right)\end{array}\right.,(j=1,2,...,L)---\left(2\right)$

In formula, L is a group number；J is the group serial number of velocity of longitudinal wave x and shear wave velocity y；Xj and yj are velocity of longitudinal wave x and horizontal stroke respectively J-th element of wave velocity y；Max (x) is the maximum for asking for velocity of longitudinal wave x；Max (y) is the maximum for asking for shear wave velocity y Value；

4) initial value φ 0 (x) and θ 0 (y) of velocity of longitudinal wave x and shear wave velocity y optimal mappings are calculated as the following formula respectively：

$\left\{\begin{array}{c}{\mathrm{\φ}}^0\left(x_j\right)=0\\ {\mathrm{\θ}}^0\left(y_j\right)=y_j\end{array}\right.,\left(j=1,2,\mathrm{...},L\right)---\left(3\right)$

In formula, L is the group number of velocity of longitudinal wave x and shear wave velocity y；J is the group serial number of velocity of longitudinal wave x and shear wave velocity y；Xj and Yj is j-th element of velocity of longitudinal wave x and shear wave velocity y respectively；

5) optimal mapping φ k (xj) and θ k (yj) of velocity of longitudinal wave x and shear wave velocity y are asked for according to the following formula：

$\left\{\begin{array}{c}{\mathrm{\φ}}^k\left(x_j\right)=S_x\left({\mathrm{\θ}}^{k-1}\right(y_j\left)\right)\\ {\mathrm{\θ}}^k\left(y_j\right)=S_y\[{\mathrm{\φ}}^k\left(x_j\right)\]/\left|\right|S_y\[{\mathrm{\φ}}^k\left(x_j\right)\]\left|\right|\end{array}\right.,(j=1,2,...,L)---\left(4\right)$

In formula, k is iterationses；S refers to and data is smoothed that smoothing method is using splines method or super smoothing method； Xj and yj are j-th element of velocity of longitudinal wave x and shear wave velocity y respectively；φ k (xj) and θ k (yj) are kth time iteration respectively, indulge Wave velocity x and j-th element of shear wave velocity y optimal mappings φ k (x) and θ k (y)；

6) iteration error is calculated as follows：

$e^2\[{\mathrm{\θ}}^k\left(y_j\right),{\mathrm{\φ}}^k\left(x_j\right)\]=\frac{1}{L}\underset{j=1}{\overset{L}{\Σ}}{\[{\mathrm{\θ}}^k\left(y_j\right)-{\mathrm{\φ}}^k\left(x_j\right)\]}^2---\left(5\right)$

In formula：K is iterationses；

7) repeat step 5) and 6), until adjacent iteration error twice meets following formula：

|e²[θ^k(yj),φ^k(xj)]-e²[θ^k-1(yj),φ^k-1(xj)]|≤δ (6)

In formula, k is iterationses；δ is error threshold value, and span is 0.001 to 0.01；When adjacent iteration error twice is expired When sufficient formula (6), kth time iteration obtains the optimal mapping function that θ (y) is referred to as shear wave velocity y, and the φ (x) for obtaining is referred to as compressional wave The optimal mapping function of speed x；φ^kAnd θ (xj)^k(yj) it is kth time iteration respectively, velocity of longitudinal wave x and shear wave velocity y most preferably becomes Change φ^k(x) and θ^kJ-th element of (y)；

8) by step 7) functional value of optimal mapping function # (y) that obtains and φ (x), returned using linear regression method, Obtain the functional relation of θ (y) and φ (x)：

θ (y)=f (φ (x)) (7)

In formula, f is the functional relation of θ (y) and the φ (x) obtained by linear regression method；

9) according to functional relation θ (y), φ (x) and f is calculated the velocity of longitudinal wave x or shear wave velocity in other wells or area y.

2. method according to claim 1, feature is step 9) calculating process be：First the vertical of other wells or area Wave velocity x or shear wave velocity y substitute into functional relation φ (x) or θ (y), obtain the optimal mapping or shear wave velocity of velocity of longitudinal wave x The optimal mapping of y, asks for the optimal mapping of the optimal mapping or velocity of longitudinal wave x of shear wave velocity y, last root further according to formula (7) According to functional relation θ (y) reverse shear wave velocity y or according to functional relation φ (x) reverse velocity of longitudinal wave x.