A method for correcting seismic layer velocity field

Abstract

The present invention relates to a seismic layer velocity field correction method, belonging to the technical field of oil and gas field exploration. The method first calculates a normalized coherence body of seismic data, then determines a region to be corrected by a coherence value, and finally corrects the seismic layer velocity field in a cluttered region by using a multi-control correction equation, thereby solving the problem of inaccurate velocity field modeling in cluttered data regions such as low signal-to-noise ratio regions and fracture regions.

Description

Technical Field

[0001] The invention relates to the technical field of oil and gas field exploration, and in particular to a seismic layer velocity field correction method. Background Art

[0002] Seismic layer velocity refers to the longitudinal wave velocity of each underground layer, which plays an important role in the processing and interpretation of seismic exploration, such as depth migration, reservoir prediction, pressure prediction, etc. At the same time, with the development of oil and gas exploration, the accuracy requirements of seismic layer velocity field in seismic exploration are increasing, and it is often necessary to establish a velocity field that can accurately describe the layer velocity of any sampling point underground.

[0003] At present, the mainstream methods for establishing layer velocity fields can be divided into two categories: (1) stacking velocity conversion: seismic processing will form a stacking velocity field similar to the root mean square velocity during stacking and migration, and the layer velocity field can be obtained through formula conversion; (2) seismic inversion: another commonly used method is to perform seismic inversion based on wave impedance, quantitatively calculate the elastic information of the seismic body response, and thus obtain the layer velocity field. These two methods have a sufficient theoretical basis and methodological technology in the process of continuous development, such as introducing neural networks in stacking velocity conversion, improving the accuracy and density of time-velocity sequence picking, thereby obtaining a more accurate velocity field, or using various inversion methods in seismic inversion.

[0004] However, in areas with cluttered seismic data (low signal-to-noise ratio areas, fault areas), the seismic wave reflection is complex and the imaging quality is poor, so the layer velocity fields established by these two methods are still inaccurate. In the picking of stacking velocity, these areas often do not have focusing in the velocity spectrum, resulting in unreliable picking results; in seismic inversion, due to the cluttered data and low credibility of amplitude information, the inversion results often cannot represent the velocity distribution of the medium. The deviation of the layer velocity field in these areas from the true velocity of the medium is almost impossible to completely solve, and must be modified by some correction methods, but the current technical methods for this are still relatively scarce. Summary of the invention

[0005] In order to solve the problem of inaccurate layer velocity field in the cluttered area of ​​seismic data (low signal-to-noise ratio area, fault area), the present invention proposes a seismic layer velocity field correction method. The method first divides the target layer segment to be corrected into several equal-proportional stratigraphic slices, then calculates the coherence value of each equal-proportional slice and normalizes it, and then calculates the correction amount of the slice in the low-coherence area according to the set threshold, and finally distributes the correction amount to each sampling point in the slice, which overcomes the above-mentioned defects and can obtain good results. The specific technical scheme is as follows:

[0006] S1: Obtain the seismic records and initial layer velocity field of the target layer segment, calculate the coherence volume of the target layer segment, and normalize the coherence value C to 0-10.

[0007] S2: Divide the coherence volume and the layer velocity field into several equal-proportion stratum slices, and use the data V in the slices as the 0 , C 0 The average value of the slice is taken as the plane value of the slice

[0008] S3: According to the set coherence threshold C t , judge the high and low coherence areas, for the low coherence area The correction equation is used to calculate the correction amount of the layer velocity, and the correction amount is applied to the data of each point in the proportional slice to obtain the correction result V 1 ; The correction equation is as follows,

[0009]

[0010] In the formula, is the slice velocity of the point to be corrected; is the standard speed near the point, which is calculated by searching for speeds with coherence values ​​greater than the coherence threshold within a fixed radius of the point and finding the average of these speeds V w is the well model value of the point, and the calculation method is

[0011]

[0012] Where N is the number of wells within a fixed radius of the point, r i is the distance from the point to the well, V wi is the well point velocity obtained by the same proportional slicing method of the well curve; a, b, c are the proportions of these three velocities in the correction amount, and the calculation method is,

[0013] c=(1-ab),

[0014] in is the coherence value of the point. a and b both reflect the deviation of the speed of the point from the standard speed. When the deviation is small, Close to C t , a tends to 0, b tends to 1, and the equation Make the main contribution, c represents a normalization item, making the result reasonable;

[0015] S5: For high coherence areas It is considered that no correction is required in this area, V 1 =V 0 ;

[0016] S6: After all slices are corrected according to this scheme, the slices are combined to obtain a corrected layer velocity field.

[0017] Innovation description: The core innovation of this invention is to innovatively propose a calculation equation for interval velocity correction using well-seismic combination and multiple control.

[0018] c=(1-ab), the equation distributes the correction amount to three sub-items, which respectively reflect the reaction of the modeled velocity at the calculation point to the true value, the reaction of the surrounding velocity field to the true value of the point (the velocity change of the medium is continuous), and the trend characteristics of the well response. The coefficient of each sub-item reflects the magnitude of each influence. The equation performs well in practical applications, the correction results are reasonable, and the consistency of well-seismic data is significantly improved. A reasonable, reliable and easy-to-calculate correction method is proposed in this field. No similar research results have been seen so far.

[0019] At the same time, the present invention also proposes a method of judging unreliable areas of layer velocity fields based on normalized coherence volumes, which can accurately identify cluttered areas of seismic data. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a technical flow chart of the present invention;

[0021] Figure 2 It is a depth structure map obtained by using the velocity field of the pre-corrected layer to change the velocity;

[0022] Figure 3 The depth structure map of the velocity field of the layer after correction by this method is formed;

[0023] Table 1 is a comparison table of depth errors of the structural map at the well point before and after the interval velocity field correction. DETAILED DESCRIPTION

[0024] The following is further described in conjunction with the accompanying drawings and embodiments. The specific steps of the present invention are as follows:

[0025] Step 1: Determine the target layer segment according to the seismic horizon, and obtain the seismic records of the target layer segment and the velocity field V of the layer to be corrected. 0 ;

[0026] Step 2: Calculate the coherence volume C of the target layer segment and normalize the coherence value C to 0-10. The normalized coherence volume is recorded as C 0 ;

[0027] Step 3: normalize the coherence volume C of the target layer segment 0 and the velocity field V of the layer to be corrected 0 The seismic data volume is divided into several equal-proportion stratigraphic slices in the vertical direction. The number of slices is determined by the time thickness of the target layer segment. The slice thickness is controlled to be any value between 10 and 20 ms. The seismic data volume is sampled at one point every 2 ms. The data V at each point in the vertical direction of the slice is 0 , C0 The average value of the slice is taken as the plane value of the slice

[0028] Step 4: Set the coherence threshold C t and retrieve the first stratum slice, for The correction equation is used to calculate the correction value ΔV of the layer velocity at the data point.

[0029]

[0030] In the formula, is the standard speed of the point, which is calculated by taking the average speed of the data points with coherence values ​​greater than the coherence threshold within a fixed radius, such as 100m. w is the well model value at that point,

[0031]

[0032] Where N is the number of wells within a fixed radius of the point, such as 2000m, and r i is the distance from this point to each well, V wi is the well point velocity obtained by the same slicing method as step 3. a, b, and c are the proportions of these three velocities in the correction amount. The calculation method is:

[0033] c=(1-ab),

[0034] in is the normalized coherence value C of each point in the vertical direction of the slice 0 The average value of

[0035] Step 5: Apply the correction ΔV to the data of the slice in the longitudinal direction at this point, V 1 =V 0 +ΔV, complete the velocity correction of each data at this point in the slice;

[0036] Step 6, for data point, directly let V 1 =V 0 ;

[0037] Step 7, repeat steps 4 to 6 until the layer velocity correction of all slices is completed, and the corrected layer velocity field V is obtained. 1 ;

[0038] Step 8: Based on the layer velocity field V before and after correction 0 、V 1 Prepare depth structure maps respectively, compare the errors of the two in the well, and evaluate the correction results. Figure 2 , 3As shown in Table 1, after the layer velocity correction of this method, the accuracy of the depth structure map at the target interface is significantly improved, and the depth error is reduced to less than 10m.

[0039] Table 1

[0040]

Claims

1. A method for correcting seismic layer velocity field comprises the following specific steps: Step 1, determine the target layer segment according to the seismic layer, and obtain the seismic record of the target layer segment and the velocity field V0 of the layer to be corrected; Step 2, calculate the coherence volume C of the target layer segment, and normalize the coherence value C to 0-10, and record the normalized coherence volume as C0; Step 3: Divide the normalized coherence volume C0 of the target layer and the velocity field V0 of the layer to be corrected into several equal-proportion stratigraphic slices in the vertical direction. The number of slices is determined by the time thickness of the target layer. The slice thickness is controlled to be any value between 10 and 20 ms. The seismic data volume is sampled at a sampling point every 2 ms. The average value of the data V0 and C0 at each point in the vertical direction of the slice is taken as the plane value of the slice. Step 4: Set the coherence threshold C t and retrieve the first stratum slice, for The correction equation is used to calculate the correction value ΔV of the layer velocity at the data point. In the formula, is the standard speed of the point, which is calculated by taking the average speed of the data points with coherence values ​​greater than the coherence threshold within a fixed radius of 100m. w is the well model value at that point, Where N is the number of wells within a fixed radius of the point, the radius value is set to 2000m, r i is the distance from this point to each well, V wi is the well point velocity obtained by the same slicing method as step 3. a, b, and c are the proportions of these three velocities in the correction amount. The calculation method is: in is the average value of the normalized coherence value C0 at each point in the longitudinal direction of the slice; Step 5, applying the correction amount ΔV to the data of the slice at the point in the longitudinal direction, V1=V0+ΔV, to complete the speed correction of each data of the slice at the point; Step 6, for For the data point, directly set V1 = V0; Step 7, repeating steps 4 to 6 until the slice velocity correction of all slices is completed to obtain the corrected slice velocity field V1; Step 8, prepare depth structure maps based on the interval velocity fields V0 and V1 before and after correction, compare the errors of the two on the well, and evaluate the correction results.

Images