A high-resolution processing method based on multi-band phase control compensation

Abstract

The present invention proposes a high-resolution processing method based on multi-band phase-controlled compensation. The method divides a seismic data body into seismic data bodies of three frequency bands: low, medium and high. Then, in each frequency band, waveform classification is performed using different seismic response characteristics of underground geological structures along the horizontal direction. The seismic data bodies are divided into different seismic phases. Representative wells of different waveform categories are selected, and synthetic records are convolved with corresponding frequency band sub-waves. The synthetic records and wellside seismic trace records after compensation are analyzed based on the minimum difference principle, and the compensation operator is calculated. The concept of seismic phase is introduced to control the corresponding change of the compensation operator in the horizontal direction. The waveform value and the compensation operator are confirmed according to the plane position of the seismic trace of the seismic data, and compensation processing is performed to obtain seismic data bodies after compensation processing of each frequency band, and finally they are merged into a seismic data body after high-resolution processing.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas seismic exploration, and in particular to a high-resolution processing method based on multi-band phase control compensation. Background Art

[0002] As the geological structure of oil exploration and development areas becomes more complex, the exploration targets are gradually changing from thick reservoirs to thin reservoirs, so high-resolution data volumes are needed to improve the recognition rate of thin layers. When the main frequency of the target layer in the study area is low, the frequency band distribution is not ideal, and the phase axis of the target layer segment is discontinuous, it will make it difficult to effectively track the target layer and interpret it in detail. Therefore, in order to improve the resolution of seismic data and smoothly carry out subsequent seismic data processing, it is necessary to perform high-resolution compensation processing on seismic data to reasonably increase the main frequency and widen the bandwidth.

[0003] In the years of development of seismic high-resolution technology, research on well control high-resolution has emerged. It uses a method that combines well logging acoustic waves, density and other data with wellside seismic trace data for integrated joint analysis, and performs well control high-resolution processing based on methods such as deconvolution, anti-Q compensation, and amplitude compensation. Graves J (2008) used the well-controlled deconvolution technology of VSP downlink wave extraction to overcome the influence of surface conditions, propagation path and attenuation of overlying strata on seismic wavelet estimation; Wang Xianbin (2008) used VSP well-controlled seismic data processing to obtain Q factor, which increased the accuracy of the description of the inner zone; Zheng Fang (2022) applied well-controlled high-resolution technology in Bamianhe area and proved that the technology has good effect in preserving low-frequency information and expanding the frequency band; Kang Youyuan (2013) used the well-controlled SBD prediction deconvolution technology to show that the frequency band is significantly widened and has a certain effect in improving the characteristics of wave groups; Cui Yongfu (2015) proposed the use of well-controlled spectral constraint deconvolution method and designed spectral constraint operator, which more effectively limited high and low noise than conventional deconvolution to protect effective signals; well-controlled high-resolution technology introduces high-frequency information in logging acoustic waves, density and other data into seismic data to improve its resolution. The development of well-controlled high-resolution technology has a huge positive effect on seismic data processing.

[0004] Based on sufficient research and actual data application, the present invention believes that the existing well control high-resolution processing technology has at least the following problems in improving the resolution of seismic data:

[0005] 1. Optimal solution of operators: Although conventional well control high-resolution technology has achieved results in improving resolution, it often ignores the lateral changes in seismic data in the work area and the different requirements for frequency extension of different frequency band components in the process of operator calculation; when solving the operator, the overall characteristics of the logging data are applied to the entire area, without considering the lateral changes in underground geological conditions. The same operator should not be used for data operators of different seismic phases and frequency bands;

[0006] 2. Limitations of compensation processing: Most well control high-resolution technologies are processed in the full frequency band. When it is necessary to highlight low-frequency information or information within a certain frequency band, the effect of this type of full-band well control high-resolution compensation processing is usually not ideal, and lacks the flexibility to increase the main frequency and expand the bandwidth of low, medium and high frequency bands. Summary of the invention

[0007] In order to solve the defects of the prior art that the lateral changes of seismic data in the study area cannot be well considered and the compensation processing is less targeted, the present invention proposes a high-resolution processing method based on multi-band phase control compensation. The method is to divide the seismic data volume into seismic data volumes of three frequency bands: low, medium and high. Then, in each frequency band, waveform classification is performed using different seismic response characteristics of underground geological structures along the horizontal direction, and the data are divided into different seismic phases. Representative wells of different waveform categories are selected according to the plane position of the wells, and synthetic records are convolved with the corresponding frequency band sub-waves. Then, the synthetic records and the wellside seismic trace records after compensation processing are analyzed based on the minimum difference principle, and the compensation operator is calculated. The concept of seismic phase is introduced to control the corresponding change of the compensation operator in the horizontal direction. The waveform value and the compensation operator are confirmed according to the plane position of the seismic trace of the seismic data, and compensation processing is performed to obtain the seismic data volume after compensation processing of each frequency band. Finally, the seismic data volume is merged into a seismic data volume to realize the compensation processing of the target layer section in the study area. Compared with the conventional method, the method before and after the application of the method of the present invention takes into account the processing in multiple frequency bands and the lateral change nature of seismic data, and well solves the limitations of the singleness of the operator and the local frequency band compensation effect, overcomes the above defects, and achieves better results.

[0008] The specific technical solutions are as follows:

[0009] S1: First, the seismic data of the target layer in the study area are processed and divided into three frequency bands: low, medium and high;

[0010] S2: bandpass filtering is performed on the seismic data of the target layer in the study area using the three frequency bands in step 1 to obtain three arrays in the time domain, and then seismic waveform classification analysis is performed on them to obtain corresponding waveform classification arrays. The number of waveforms in the waveform classification is 3;

[0011] S3: For low frequency band:

[0012] S3-1: Input all the completed wells in the study area, determine the waveform category to which the completed wells belong according to their planar positions, and select representative wells of three types of waveforms, i.e., three representative wells.

[0013] S3-2: For the representative wells in step 3-1, synthesize the seismic records of the target layer based on the well logging acoustic wave and density data, and perform Fourier transform to obtain the frequency domain array.

[0014] S3-3: For the representative wells in step 3-1, the seismic trace data of the target layer segment are extracted respectively, and bandpass filtering is performed to obtain the frequency domain array of the seismic trace data of the representative wells in the frequency band.

[0015] S3-4: Calculate the phase control compensation operator for this frequency band:

[0016] S3-4-1: For the representative wells of the first type of waveform, the compensation operator of the waveform is calculated by the least square method based on the principle of minimizing the difference between the frequency domain array of the compensated wellside seismic trace data and the frequency domain array of the synthetic record.

[0017] S3-4-2: For the second and third type waveforms, use step 3-4-1 to calculate the respective compensation operators. S3-5: Perform compensation processing on the seismic data of the target layer section in the frequency band study area.

[0018] S3-5-1: Extract the first seismic data of the target layer in the study area, and determine the waveform value of the current seismic trace, i.e., the waveform class, based on the plane position of the seismic trace and the waveform classification array.

[0019] S3-5-2: Determine the compensation operator of the seismic trace according to the waveform value in step 3-5-1.

[0020] S3-5-3: Perform inverse Fourier transform on the compensation operator to obtain the time domain compensation operator.

[0021] S3-5-4: Using the compensation operator of step 3-5-3 and the convolution method, the seismic data array of the target layer segment of the current seismic trace is compensated to obtain the compensated seismic data array.

[0022] S3-5-5: Replace the seismic channel and repeat steps 3-5-1 to 3-5-4 until all seismic channels in the study area are compensated to obtain the compensated seismic data array of the frequency band;

[0023] S4: repeat step 3 to calculate the compensated seismic data arrays of the middle frequency band and the high frequency band;

[0024] S5: For the seismic data array after compensation of each frequency band, the inverse distance weighted method is used to fuse the arrays after compensation of the low, medium and high frequency bands to reconstruct a full-band seismic data array, which is the final result after compensation processing of the seismic data of the target layer in the study area.

[0025] Compared with the prior art, the present invention can achieve the following effects:

[0026] 1. Operator advantages: ① The concept of seismic phase is introduced, taking into account the lateral changes of the geological structure in the study area, the seismic data in the study area are divided into seismic phases according to the waveform category, and the operator is obtained according to the well logging data and seismic data of different seismic trace plane positions and waveform classifications; ② Multi-band operators are considered at the same time to calculate operators, and the operators are more subdivided and processed more purposefully; the operators involved in the compensation processing process can better reflect the seismic data of specific areas and frequency band components;

[0027] 2. Targeted compensation processing: The present invention takes into account the limitations of existing well control high-resolution technology and performs multi-band phased deconvolution calculations for different frequency bands. Compared with conventional technologies, it can better improve the low-frequency band and widen the frequency band. The frequency division takes into account that different frequency bands may require different degrees of signal compensation. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is the principle and flow chart of the present invention;

[0029] Figure 2 This is the spectrum analysis diagram of the 2500-2800ms seismic data of the target layer in a certain study area in the Sichuan Basin without processing;

[0030] Figure 3 This is a spectrum analysis diagram of 2500-2800ms seismic data of a target layer section in a certain research area of ​​Sichuan Basin after being processed by the present invention;

[0031] Figure 4 This is a well-seismic calibration map made when the 2500-2800ms seismic data of the target layer section in a certain study area in the Sichuan Basin was not processed;

[0032] Figure 5 The well-seismic calibration map is made by processing the 2500-2800ms seismic data of the target layer section in a certain research area of ​​the Sichuan Basin using the present invention. DETAILED DESCRIPTION

[0033] Example 1

[0034] A high-resolution processing method based on multi-band phase control compensation, the steps comprising:

[0035] Step 1: Input the seismic data S(x,y,t) of the target layer in the three-dimensional study area, where x represents the line number, y represents the channel number, t represents the time, the start time of the target layer is t0, and the end time is t1. Perform frequency analysis on S(x,y,t), and use the equal division method to divide it into three frequency bands: low, medium, and high according to the main frequency fmHz of the seismic data. The frequency range of the low frequency band is 0Hz to fm / 2Hz, the frequency range of the medium frequency band is fm / 2Hz to 3fm / 2Hz, and the frequency range of the high frequency band is 3fm / 2Hz to 2fmHz.

[0036] Step 2: Use the three frequency bands in step 1 to perform bandpass filtering on S(x,y,t) to obtain three arrays in the time domain: S1(x,y,t), S2(x,y,t), and S3(x,y,t); then perform seismic waveform classification analysis on them respectively, setting the number of waveform classifications to 3, and obtain the corresponding waveform classification arrays BS1(x,y,t), BS2(x,y,t), and BS3(x,y,t);

[0037] Step 3: For the low frequency band, the frequency band range is 0Hz to fm / 2Hz,

[0038] Step 3-1: Input all the completed wells in the target layer of the study area. According to the plane position of the completed wells, combined with the waveform classification array BS1(x,y,t) of the frequency band in step 2, the completed wells are divided into three categories, corresponding to three types of waveform characteristics, and representative wells of the three waveform types are selected, recorded as W1(t), W2(t), and W3(t).

[0039] Step 3-2: For the three completed wells in step 3-1, input the velocity and density logging data of the completed wells in the target layer section t0 to t1 in the study area, use the wavelet of the frequency band for convolution calculation, and obtain the synthetic records R1(t), R2(t), and R3(t). Perform Fourier transform on them to obtain the frequency domain arrays FR1(w), FR2(w), and FR3(w).

[0040] Step 3-3: Extract the wellside seismic trace data P1(t), P2(t), and P3(t) in the target layer section t0 to t1 of the three wells completed in step 3-1, and use the frequency band range parameter of the frequency band, 0Hz to fm / 2Hz, to perform frequency domain bandpass filtering to obtain the filtered frequency domain arrays FP1(w), FP2(w), and FP3(w).

[0041] Step 3-4: Calculate the phase compensation operator for this frequency band.

[0042] Step 3-4-1: For the representative well of the first type of waveform, based on the principle of minimizing the difference between the frequency domain array of the compensated wellside seismic trace data and the frequency domain array of the synthetic record, the compensation operator Q1(w) of the waveform is calculated using the least squares method.

[0043] min{Q1(w)FP1(w)-FR1(w)},

[0044] Step 3-4-2: For the second and third type waveforms, use step 3-4-1 to calculate the respective compensation operators, Q2(w) and Q3(w).

[0045] Step 3-5: Compensate the seismic data S1 (x, y, t) in the target layer segment t0 to t1 of the three-dimensional study area of ​​the frequency band.

[0046] Step 3-5-1: Extract the first seismic data M(t) of the target layer in the study area, and determine the waveform value X of the current seismic trace based on the plane position of the seismic trace and the waveform classification array, that is, the waveform class. The value range of X is 1, 2, 3.

[0047] Step 3-5-2: Determine the compensation operator Q(w) according to the value of X in step 3-5-1.

[0048] X=1, Q(w)=Q1(w); X=2, Q(w)=Q2(w); X=3, Q(w)=Q3(w),

[0049] Step 3-5-3: Perform inverse Fourier transform on Q(w) to obtain the time domain compensation operator q(t).

[0050] Step 3-5-4: Compensate the target layer segment array M(t) of the current seismic trace to obtain the compensated array Y(t).

[0051] Y(t)=M(t)*q(t), * represents convolution operation

[0052] Step 3-5-5: Replace the seismic channel and repeat steps 3-5-1 to 3-5-4 until all seismic channels in the study area are compensated. The compensated seismic data array is recorded as QS1(x, y, t);

[0053] Step 4: Repeat step 3, and perform compensation processing on the time domain seismic data arrays S2(x, y, t) and S3(x, y, t) of the mid-frequency band and high-frequency band. The process is the same as that of the low-frequency band, and the compensated seismic data arrays QS2(x, y, t) and QS3(x, y, t) are calculated.

[0054] Step 5: The three-dimensional seismic data arrays QS1(x, y, t), QS2(x, y, t), and QS3(x, y, t) of three different frequency bands after compensation processing are fused and reconstructed into a full-band three-dimensional seismic data array QS(x, y, t), i.e., the seismic data of the target layer segment in the study area after compensation processing.

[0055] Example 2

[0056] The method of the present invention was applied to the compensation processing of seismic data of 2500-2800ms in the target layer section of a certain research area in the Sichuan Basin. Figure 2 and Figure 3 The frequency spectrum analysis diagram of the 2500-2800ms seismic data of the target layer before and after the application of the present invention is shown. It can be seen that after the application of the present invention, the main frequency is improved, the bandwidth is widened, and the increase in amplitude intensity is also in line with the law; Figure 4 and Figure 5 The following is a diagram showing the effect of well seismic calibration of the target layer segment of 2500-2800ms before and after the application of the present invention. It can be seen that after the application of the present invention, the synthetic seismic records of the forward modeling of the logging data have strong consistency with the wave group characteristics of the wellside seismic trace records, high correlation coefficient, and greatly enhanced weak reflection signals. A more continuous isotropic axis appears, and the compensation effect is significant.

[0057] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A high-resolution processing method based on multi-band phase control compensation includes the following specific steps: Step 1, first input the seismic data S(x,y,t) of the target layer segment in the three-dimensional study area, where x represents the line number, y represents the channel number, t represents the time, the start time of the target layer segment is t0, and the end time is t1, and perform frequency analysis on S(x,y,t). According to the main frequency fmHz of the seismic data, the data is divided into three frequency bands: low, medium and high. The frequency range of the low frequency band is 0Hz to fm / 2Hz, the frequency range of the medium frequency band is fm / 2Hz to 3fm / 2Hz, and the frequency range of the high frequency band is 3fm / 2Hz to 2fmHz. Step 2: bandpass filter S(x, y, t) using the three frequency bands in step 1 to obtain three arrays in the time domain: S1(x, y, t), S2(x, y, t), and S3(x, y, t); then perform seismic waveform classification analysis on them respectively, setting the number of waveform classifications to 3, and obtain the corresponding waveform classification arrays BS1(x, y, t), BS2(x, y, t), and BS3(x, y, t); Step 3, for the low frequency band, the frequency band range is 0Hz to fm / 2Hz, Step 3-1: Input all the completed wells in the target layer of the study area. According to the plane position of the completed wells, combined with the waveform classification array BS1(x, y, t) of the frequency band in step 2, the completed wells are divided into three categories, corresponding to three waveform characteristics, and representative wells of the three waveform types are selected, recorded as W1(t), W2(t), and W3(t). Step 3-2: For the three completed wells in step 3-1, input the velocity and density logging data of the completed wells in the target layer section t0 to t1 in the study area, use the wavelet of the frequency band to perform convolution calculation, obtain the synthetic records R1(t), R2(t), R3(t), perform Fourier transform on them to obtain the frequency domain arrays FR1(w), FR2(w), FR3(w), Step 3-3, extract the wellside seismic trace data P1(t), P2(t), P3(t) in the target layer section t0 to t1 of the three wells completed in step 3-1, use the frequency band range parameter of the frequency band, 0Hz to fm / 2Hz, perform frequency domain bandpass filtering, and obtain the filtered frequency domain arrays FP1(w), FP2(w), FP3(w), Step 3-4, calculate the phase control compensation operator of the frequency band, Step 3-4-1, for the representative well of the first type of waveform, based on the principle of minimizing the difference between the frequency domain array of the compensated wellside seismic trace data and the frequency domain array of the synthetic record, the compensation operator Q1(w) of the waveform is calculated using the least squares method. min{Q1(w)FP1(w)-FR1(w)}, Step 3-4-2, for the second and third type waveforms, use step 3-4-1 to calculate the respective compensation operators, Q2(w) and Q3(w), Step 3-5, compensating the seismic data S1 (x, y, t) in the target layer segment t0 to t1 of the three-dimensional study area of ​​the frequency band, Step 3-5-1, extract the first seismic data M(t) of the target layer segment in the study area, and determine the waveform value X of the current seismic trace based on the plane position of the seismic trace and the waveform classification array, that is, the waveform class, and the value range of X is 1, 2, 3. Step 3-5-2, determine the compensation operator Q(w) according to the value of X in step 3-5-1, X=1, Q(w)=Q1(w); X=2, Q(w)=Q2(w); X=3, Q(w)=Q3(w), Step 3-5-3, perform inverse Fourier transform on Q(w) to obtain the time domain compensation operator q(t), Step 3-5-4, compensate the target layer segment array M(t) of the current seismic trace to obtain the compensated array Y(t). Y(t)=M(t)*q(t), * represents convolution operation Step 3-5-5, replace the seismic channel, repeat steps 3-5-1 to 3-5-4 until all seismic channels in the study area are compensated, and the seismic data array after compensation is recorded as QS1(x, y, t); Step 4, repeat step 3, and perform compensation processing on the time domain seismic data arrays S2(x, y, t) and S3(x, y, t) of the mid-frequency band and high-frequency band. The process is the same as that of the low-frequency band, and the compensated seismic data arrays QS2(x, y, t) and QS3(x, y, t) are calculated; Step 5, the three-dimensional seismic data arrays QS1(x, y, t), QS2(x, y, t), and QS3(x, y, t) of three different frequency bands after compensation processing are fused and reconstructed into a full-band three-dimensional seismic data array QS(x, y, t), i.e., the seismic data of the target layer segment in the study area after compensation processing.

Images