Method for enhancing geological object in frequency domain based on target orientation
By transforming seismic attribute data from the time domain to the Z-frequency domain for target-oriented shaping analysis, the problem of noise interference in seismic data was solved, resulting in a significant enhancement of geological targets and improved interpretation accuracy.
Patent Information
- Application Number
- CN202410475343.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-19
- Publication Date
- 2025-10-24
AI Technical Summary
Current seismic exploration methods contain a lot of noise in seismic data, which increases the difficulty of seismic interpretation and makes it difficult to directly distinguish deeply buried geological targets from their surrounding geological environment through seismic attributes.
Seismic attribute data is transformed from the time domain to the Z-frequency domain, and target-oriented shaping analysis is performed to preserve and enhance the spectral morphology of geological targets. The spectral characteristics of geological targets are further enhanced through artificial intervention, and finally transformed back to the time domain.
It effectively distinguishes geological targets from their surrounding geological environment, improving the accuracy of seismic interpretation and the prominence of geological targets.
Smart Images
Figure BDA0004800330700000021 
Figure BDA0004800330700000031 
Figure BDA0004800330700000032
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of oil and gas exploration interpretation, and in particular relates to a method for enhancing geological objects in the frequency domain based on target orientation. Background Art
[0002] Seismic data acquired using current seismic exploration methods contain significant amounts of systematic and random noise. Current seismic exploration methods primarily rely on the propagation of elastic waves in the Earth's crust, generated by artificial means such as explosions or impacts, and the recording of reflected and refracted waves detected by geophones. During seismic data acquisition, the data often contain significant noise due to the influence of numerous complex factors, including the earthquake source, the excitation method, the medium characteristics of the excitation point, the receiving environment, and the surface and subsurface structures. This complicates seismic interpretation. Seismic attributes, as an observational perspective on seismic data, can highlight specific aspects of the data. Seismic reflections originate from subsurface strata. Lateral variations in subsurface strata characteristics lead to lateral variations in seismic reflection characteristics, which in turn influence seismic attribute variations. Therefore, seismic attributes carry information about subsurface strata, providing the physical basis for using seismic attributes to predict geological target parameters. However, geological targets are deeply buried, blending seamlessly with the surrounding geological environment, and, coupled with noise interference, are difficult to distinguish directly using seismic attributes in the airspace. Summary of the Invention
[0003] To overcome the shortcomings of the prior art, the present invention provides a method for enhancing geological objects in the frequency domain based on target orientation. First, the seismic attribute volume is transformed from the time domain to the Z frequency domain. Then, the seismic attribute spectrum is subjected to target-oriented shaping analysis to discover the association between the geological target object and the spectrum. The spectrum form that can best enhance the geological target object is retained. Finally, the shaped frequency domain data is inversely transformed back to the time domain to achieve the purpose of enhancing the geological target object and making it significantly distinguishable from the surrounding geological environment.
[0004] The above object of the present invention is achieved through the following technical solution: a method for enhancing geological objects in the frequency domain based on target orientation, comprising the following steps:
[0005] 1. Select the pre-studied seismic attribute data volume, perform corresponding denoising and energy equalization, and linearly transform the attribute values to values between -1 and 1;
[0006] 2. Transform seismic attribute data from time domain data to Z domain data through channel-by-channel z-transformation;
[0007] 3. Select geological target samples: Manually mark the typical area of geological targets on the time domain data volume section, and set the time domain data outside this area to 0, while the data within the area remains unchanged;
[0008] 4. Perform z transform on each trace of the geological target sample to obtain Z-domain data of the geological target sample;
[0009] 5. Perform averaging processing on the Z-domain data of all geological target samples to obtain average Z-domain data of the geological target sample;
[0010] 6. Fuse the attribute body Z-domain data and the average Z-domain data of the geological target sample. Set a ratio coefficient artificially, convert the amplitude of the average Z-domain data of the target sample into a coefficient greater than 1, and multiply the coefficient with each trace Z-domain data of the attribute body to obtain enhanced attribute body Z-domain data;
[0011] 7. Perform inverse z transform on the enhanced attribute body Z-domain data to obtain time-domain data body of the geological target enhancement, and linearly transform the attribute value back to the original value range.
[0012] Further, the seismic attribute data body in step 1 is a three-dimensional array, denoted as D LTS , where L is the line number, T is the number of traces per line, and S is the number of sampling points per trace, denoted as D; D(l, t, s) is a sampling point data of D, where l is the line index, t is the trace index, and s is the sampling point index, 0 = < l < L, 0 = < t < T, 0 = < s < S, l, t, s are all integers; D(l, t) is the sampling data of the lth line and the tth trace of D, and D(l) is the sampling data of the lth line of D.d max , d min The linear transformation method of the attribute value of the seismic attribute data body is as follows: sequentially transform each sampling point data of D, and the formula is as follows:
[0013] d l,t,s = D(l, t, s)
[0014]
[0015] Further, the z transform of each trace in step 2 is performed as follows:
[0016] Sequentially transform each trace sampling data of D, and the formula is as follows:
[0017] d l,t = D(l, t)
[0018]
[0019] f(z) is the Z-domain data of the data trace d l,t .
[0020] Further, the formula used in step 3 is as follows:
[0021]
[0022] When s is in the region:
[0023] When s is out of the region:
[0024] Further, the formula used in step 4 is as follows:
[0025]
[0026] f1(z) is the Z-domain data of the target sample data channel f2(z) is the Z-domain data of the target sample data channel
[0027] Further, the formula used in step 5 is as follows:
[0028] f a (z) = (f1(z) + f2(z)) / 2
[0029] f a (z) is the average Z-domain data of the geological target sample.
[0030] Further, the formula used in step 6 is as follows:
[0031] f d (z) = f(z)·(1+α·f a (z))
[0032] Wherein, α is an artificially set weight coefficient, generally taking a value between 0.05 and 0.35, f d (z) is the obtained enhanced attribute body Z-domain data.
[0033] Further, the formula used in step 7 is as follows:
[0034]
[0035] D(l,t,s) = d(s).
[0036] The beneficial effects of the present application compared with the prior art are: different from the traditional time-space domain and frequency-space domain feature enhancement method, the present application proposes a frequency domain geological target enhancement method, which combines the z transform of the local geological target with the z transform of the overall geological environment, and adds artificial intervention, so that the attribute enhancement processing is no longer blindly performed, and has an immediate effect. BRIEF DESCRIPTION OF DRAWINGS
[0037] The present application will be further described below in conjunction with the drawings and specific embodiments
[0038] Figure 1 is a schematic diagram of original seismic data;
[0039] Figure 2 is a schematic diagram of instantaneous amplitude attribute data;
[0040] Figure 3 is a schematic diagram of selecting geological target samples from the instantaneous amplitude attribute data;
[0041] Figure 4 is a histogram of the value distribution of the instantaneous amplitude attribute data;
[0042] Figure 5 is a schematic diagram of the instantaneous amplitude attribute data after low-pass filtering;
[0043] Figure 6 is a histogram of the value distribution of the instantaneous amplitude attribute data after low-pass filtering;
[0044] Figure 7 is a schematic diagram of the instantaneous amplitude attribute data after processing by the method;
[0045] Figure 8 is a histogram of the value distribution of the instantaneous amplitude attribute data after processing by the method.
[0046] Figure 9 is the present application and flow chart. DETAILED DESCRIPTION
[0047] The present application will be described in detail below with specific examples, but the scope of protection of the present application is not limited. Unless otherwise specified, the experimental methods used in the present application are conventional methods, and the experimental apparatus, materials, reagents, etc. used can be obtained from commercial channels.
[0048] Example 1
[0049] A target-oriented frequency domain method for enhancing geological objects
[0050] Step 1: Select the pre-study seismic attribute data volume, and perform corresponding denoising and energy equalization, and linearly transform the attribute values to values between -1 and 1. Figure 2 The attribute data selected for this example is instantaneous amplitude attribute data.
[0051] The seismic attribute data volume is a three-dimensional array, denoted as D LTSwhere L is the number of lines, T is the number of traces per line, S is the number of samples per trace, and D is the abbreviation. Assuming that the user has pre-processed D with denoising and energy equalization, this paper focuses on the linear transformation of the data in D to a value between -1 and 1. D(l, t, s) is a sample point data of D, where l is the line index, t is the trace index, s is the sample point index, 0 = < l < L, 0 = < t < T, 0 = < s < S, l, t, s are all integers; D(l, t) is the sample data of the lth line and the tth trace of D, and D(l) is the sample data of the lth line of D.d max 、d min
[0052] d l,t,s = D(l, t, s)
[0053]
[0054] Step two: transform the seismic attribute data from time domain data to Z domain data by trace-by-trace z-transformation.
[0055] Transform each trace sample data of D by z-transformation:
[0056] d l,t = D(l, t)
[0057]
[0058] f(z) is the Z domain data of the data trace d l,t .
[0059] Step three: select the geological target sample: manually mark the typical area of the geological target on the time domain data body profile, and set the time domain data outside this area to 0, and the data in the area remains unchanged. Figure 3 The geological target sample is selected from the instantaneous amplitude attribute data in this example.
[0060] Suppose the manually marked typical area of the geological target contains 2 traces of data
[0061]
[0062] When s is in the area:
[0063] When s is outside the area:
[0064] Step four: perform z-transformation on the geological target sample trace by trace to obtain the Z domain data of the geological target sample.
[0065]
[0066] f1(z) is the Z-domain data of the target sample trace f2(z) is the Z-domain data of the target sample trace f2(z) is the Z-domain data of the target sample trace
[0067] Step five: average the Z-domain data of all the target sample traces to obtain the average Z-domain data of the target sample traces.
[0068] f a (z) = (f1(z) + f2(z)) / 2
[0069] f a (z) is the average Z-domain data of the target sample traces.
[0070] Step six: fuse the attribute Z-domain data and the average Z-domain data of the target sample traces. Set the ratio coefficient artificially, convert the amplitude of the average Z-domain data of the target sample traces into a coefficient greater than 1, and multiply it with each trace of the attribute Z-domain data to obtain the enhanced attribute Z-domain data.
[0071] f d (z) = f(z) · (1 + a · f a (z))
[0072] wherein a is the weight coefficient set artificially, generally taking a value between 0.05 and 0.35, and f d (z) is the enhanced attribute Z-domain data.
[0073] Step seven: perform inverse Z-transformation on the enhanced attribute Z-domain data to obtain the time-domain data body of the enhanced geological target, and linearly transform the attribute value back to the original value range.
[0074]
[0075] D(l, t, s) = d(s)
[0076] If the enhancement effect is not ideal, there are three processing methods: 1. abandon the processing result of step seven, return to step six to adjust a, and then execute step seven; 2. abandon the processing results of steps four to seven, return to step three to re-construct the target sample, and then execute until step seven; 3. on the basis of the existing results, execute from step three until step seven. Figure 7 is the instantaneous amplitude attribute data processed by the method. In order to better illustrate the method, the data of Figure 2 is processed by low-pass filtering, and the effect is shown in Figure 5 . Not only from the processing effect, but also the geological target can be enhanced by the method, compared with Figure 4 , Figure 6 , Figure 8The distribution of the data processed by the method changes greatly and is closer to normal distribution, highlights the data related to the geological target, and suppresses the data irrelevant to the geological target.
[0077] The above-described embodiments are merely preferred embodiments of the present application and are not all implementable embodiments of the present application. Any obvious modifications made by those skilled in the art without departing from the principles and spirit of the present application should be considered to be within the scope of the claims of the present application.
Claims
1. A method of target-oriented frequency domain enhanced geologic object based on, characterized in that, The steps are as follows: S1. Select the pre-study seismic attribute data body, make corresponding denoising and energy equalization, and linearly transform the attribute value to the value between-1 and 1; S2. Transform the seismic attribute data from time domain data to Z domain data through z transform; S3. Select a geological target sample: manually mark the typical area of the geological target on the time domain data profile, and set the time domain data outside the area to 0, and the data in the area remains unchanged; S4. z transform is performed on the geological target sample to obtain the Z domain data of the geological target sample; S5. Average the Z domain data of all geological target samples to obtain the average Z domain data of the geological target sample; S6. Fuse the attribute body Z domain data and the average Z domain data of the geological target sample. Manually set the ratio coefficient, convert the amplitude of the average Z domain data of the target sample into a coefficient greater than 1, and multiply it by each Z domain data of the attribute body to obtain the enhanced attribute body Z domain data; S7. Reverse z transform is performed on the enhanced attribute body Z domain data to obtain the time domain data body of the enhanced geological target, and the attribute value is linearly transformed back to the original value range.
2. The target-oriented, frequency-domain enhanced geologic object-based method of claim 1, wherein, The seismic attribute data volume in the step S1 is a three-dimensional array, denoted as D LTS , wherein L is the number of lines, T is the number of traces per line, and S is the number of sampling points per trace, and D is abbreviated; D(l, t, s) is a sampling point data of D, wherein l is a line index, t is a trace index, and s is a sampling point index, 0=<l<L, 0=<t<T, 0=<s<S, l, t, and s are all integers; D(l, t) is sampling data of the lth line and the tth trace of D, and D(l) is sampling data on the lth line of D.d max , d min are respectively the maximum sampling point value and the minimum sampling point value of D, and the linear transformation method of the seismic attribute data volume attribute value is: sequentially transforming each sampling point data of D, and the formula is as follows: d l,t,s = D(l, t, s) 3. The target-oriented, frequency-domain enhanced geologic object-based method of claim 1, wherein, The specific operation of z transform in step S2 is to sequentially perform z transform on each sampling data, and the formula is as follows: d l,t = D(l, t) f(z) is the Z-domain data of the data channel d l,t .
4. The target-oriented, frequency-domain enhanced geologic object-based method of claim 1, wherein, The formula used in step S3 is as follows: When s is within the region: When s is outside the region: d l1,t1,s = 0, d l2,t2,s = 0.
5. The target-oriented, frequency-domain enhanced geologic object-based method of claim 1, wherein, The formula used in step S4 is as follows: f1(z) is the Z-domain data of the target sample trace f2(z) is the Z-domain data of the target sample trace 6. The target-oriented, frequency-domain enhanced geologic object-based method of claim 1, wherein, The formula used in step S5 is as follows: f a (z) = (f1(z) + f2(z)) / 2 f a (z) is the average Z-domain data for the geologic target sample.
7. The target-oriented, frequency-domain enhanced geologic object-based method of claim 1, wherein, The formula used in step S6 is as follows: f d (z) = f(z) · (1 + a · f a (z)) wherein a is a weight factor set artificially, generally taking a value between 0.05 and 0.35, f d (z) is to obtain enhanced attribute volume Z domain data.
8. The target-oriented, frequency-domain enhanced geologic object-based method of claim 1, wherein, The formula used in step S7 is as follows: D(l,t,s)=d(s).