Methods, systems, and computer-readable storage media for predicting a hydrocarbon source rock

Abstract

This invention relates to a method, system, and computer-readable storage medium for predicting source rocks. The method includes the following steps: First, by combining well-seismic calibration and full-area tracking, the top and bottom interfaces of source rocks in time-domain seismic data are accurately located, overcoming waveform distortion caused by variations in formation thickness. Then, time-spectrum analysis is used to extract characteristic frequencies of the source rocks, focusing the analysis on their properties and reducing irrelevant interference. Further, frequency-division interpretation is performed based on the characteristic frequencies, converting the time-domain data into the frequency domain to leverage the advantages of the frequency domain for waveform clustering analysis. Shannon interpolation is used to process the frequency domain data, improving data quality, enhancing smoothness and continuity, and suppressing noise. Finally, waveform clustering is performed to reduce the impact of waveform distortion, improve clustering stability and reliability, and thus visually display the planar distribution of source rocks, providing accurate and reliable geological data for oil and gas exploration and development.

Description

Technical Field

[0001] This invention relates to the field of hydrocarbon source rock prediction technology, and in particular to a method, system, and computer-readable storage medium for predicting hydrocarbon source rocks. Background Technology

[0002] The frequency, phase, and amplitude information of seismic data are ultimately reflected in seismic waveforms; that is, changes in seismic waveforms reflect changes in seismic signals. Based on this theoretical understanding, geophysical exploration applies waveform clustering to seismic facies research, sedimentary microfacies prediction, and lithofacies classification. The principle is vector clustering, which groups waveforms with high similarity into the same class and those with low similarity into different classes. Current waveform clustering methods primarily use "cluster centers" to represent typical waveforms of different classes. Initial cluster centers are randomly selected, and then the cluster centers are iteratively updated until they converge to the true cluster centers. Finally, the waveforms of each seismic trace are divided using the cluster centers. A waveform with the highest similarity to a particular cluster center is assigned to that class, and different waveforms represent different seismic facies. Seismic facies to sedimentary facies transitions are then achieved based on the principles of seismic stratigraphy. Most exploration workers use waveform clustering methods to predict reservoir distribution, and some experts and scholars use this method to characterize volcanic rock lithofacies. However, its application to the study of source rock distribution is a first.

[0003] Traditional waveform clustering methods mostly divide seismic phases based on the similarity of seismic trace waveforms within an equal time window. However, when the thickness of the strata varies greatly, resampling the seismic signal into an equal-length signal in the time domain can easily cause waveform distortion, which limits the application of these methods when the thickness of the target stratum varies greatly. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method, system and computer-readable storage medium for predicting source rocks, addressing the problem of inaccurate waveforms in the prior art.

[0005] The technical solution adopted by this invention to solve its technical problem is: a method for predicting hydrocarbon source rocks, comprising the following steps:

[0006] S1: Acquire time-domain seismic data and actual drilling data of the target area after stacking;

[0007] S2: Based on the actual drilling data and the time-domain seismic data, perform well-seismic joint calibration to determine the top and bottom interfaces of the source rock in the time-domain seismic data, and conduct full-area tracking based on the top and bottom interfaces in the time-domain seismic data to obtain the top and bottom surface data of the source rock.

[0008] S3: Based on the actual drilling data after well-seismic joint calibration, time-domain seismic data is analyzed using time-frequency analysis to obtain the characteristic frequencies of the source rock;

[0009] S4: Based on the characteristic frequencies of the source rock, the time-domain seismic data is interpreted by frequency division to determine the seismic frequency division data corresponding to the characteristic frequencies of the source rock;

[0010] S5: Convert the time-domain seismic frequency division data into frequency-domain seismic frequency division data, and use the Shannon interpolation method to interpolate the frequency-domain seismic frequency division data to determine the interpolated frequency-domain seismic frequency division data.

[0011] S6: Based on the top and bottom surface data of the source rock and the frequency domain seismic frequency division data after interpolation, perform waveform clustering between the top and bottom surfaces to obtain a waveform clustering planar map of the target area reflecting the planar distribution of the source rock.

[0012] In one embodiment, the actual drilling data includes acoustic data and logging density data, and S2 includes:

[0013] Based on the acoustic data and the well logging density data, velocity data and density data of the source rock are obtained;

[0014] Based on the velocity and density data of the source rock, the time-domain seismic data is jointly calibrated by well and seismic data to determine the top and bottom interfaces of the time-domain seismic data corresponding to the source rock. Full-area tracking is then carried out on the time-domain seismic data to obtain the top and bottom surface data of the source rock.

[0015] In one embodiment, the wave impedance characteristics of the source rock include a three-peak, two-valley characteristic and a two-peak, two-valley characteristic.

[0016] In one embodiment, the logging data of the actual drilled well includes sonic transit time curves and density curves, and the time-domain seismic data includes actual seismic records;

[0017] The method of performing well-seismic joint calibration of the time-domain seismic data based on the wave impedance characteristics of the source rock to determine the top and bottom interfaces of the time-domain seismic data corresponding to the source rock includes:

[0018] The acoustic transit time curve and the density curve are synthesized to obtain a synthetic seismic record.

[0019] By comparing the cross-correlation between the actual seismic records and the synthetic seismic records, the top and bottom interfaces of the time-domain seismic data corresponding to the source rocks are determined.

[0020] In one embodiment, determining the top and bottom interfaces of the time-domain seismic data corresponding to the source rock by comparing the cross-correlation between actual seismic records and synthetic seismic records includes:

[0021] Based on the top and bottom interfaces in the actual drilling data, three-dimensional seismic phase tracking technology is used to track the time-domain seismic data to determine the top and bottom interfaces of the time-domain seismic data corresponding to the source rock.

[0022] In one embodiment, S3 includes:

[0023] Based on the actual drilling data after well-seismic joint calibration, time-domain seismic data is analyzed using time-frequency analysis to obtain the characteristic frequency range of the source rock. The average value of the characteristic frequency range of the source rock is calculated to obtain the characteristic frequency of the source rock.

[0024] In one embodiment, S4 includes:

[0025] The time-domain seismic data is subjected to tectonic-guided filtering. Based on the characteristic frequencies of the source rocks, the time-domain seismic data after tectonic-guided filtering is interpreted by frequency division to determine the seismic frequency division data corresponding to the characteristic frequencies of the source rocks.

[0026] In one embodiment, converting the time-domain seismic frequency-division data into frequency-domain seismic frequency-division data includes:

[0027] The time-domain seismic frequency-division data is converted into frequency-domain seismic frequency-division data using the Fourier transform method.

[0028] The present invention also provides a system for predicting source rocks, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method for predicting source rocks described above.

[0029] The present invention also provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the method for predicting source rocks as described above.

[0030] Implementing this invention offers the following advantages: By combining well-seismic calibration and full-area tracking, the top and bottom interfaces of source rocks in time-domain seismic data are accurately located, overcoming waveform distortion caused by variations in formation thickness. Subsequently, time-spectrum analysis is used to extract characteristic frequencies of the source rocks, focusing the analysis on their properties and reducing irrelevant interference. Furthermore, frequency-division interpretation is performed based on these characteristic frequencies, converting the time-domain data into the frequency domain to leverage its advantages for waveform clustering analysis. Shannon interpolation is used to process the frequency-domain data, improving data quality, enhancing smoothness and continuity, and suppressing noise. Finally, waveform clustering is performed to reduce the impact of waveform distortion, improve clustering stability and reliability, and thus visually demonstrate the planar distribution of source rocks, providing accurate and reliable geological data for oil and gas exploration and development. Attached Figure Description

[0031] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings:

[0032] Figure 1 This is a flowchart of an embodiment of the method for predicting hydrocarbon source rocks according to the present invention;

[0033] Figure 2 This is a flowchart of another embodiment of the method for predicting source rocks according to the present invention;

[0034] Figure 3 This is a schematic diagram of well-seismic joint calibration of an embodiment of the method for predicting hydrocarbon source rocks according to the present invention;

[0035] Figure 4 This is a schematic diagram of the seismic facies characteristics of a source rock with a characteristic frequency of 25 Hz, as described in an embodiment of the method for predicting source rocks according to the present invention.

[0036] Figure 5 This is a schematic diagram of clustering features based on a 25Hz characteristic frequency waveform, representing an embodiment of the method for predicting source rocks according to the present invention. Detailed Implementation

[0037] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0038] The technical problem in the background technology is that traditional waveform clustering methods suffer from decreased clustering accuracy and limited application due to waveform distortion under complex geological structures, especially when the stratum thickness varies greatly.

[0039] This invention provides a method for predicting source rocks, which can achieve the following objectives: accurately identifying the location of source rocks in time-domain seismic data through well-seismic joint calibration and full-area tracking; extracting characteristic frequencies of source rocks using time-spectrum analysis technology to provide accurate parameters for waveform clustering; reducing waveform distortion and improving the stability and reliability of waveform clustering by employing frequency domain processing and interpolation methods; and finally performing efficient waveform clustering on the interpolated frequency-domain seismic frequency-division data to generate a waveform clustering planar map reflecting the planar distribution of source rocks, providing intuitive and accurate prediction of source rock distribution for oil and gas exploration and development, and significantly improving exploration efficiency and success rate.

[0040] Specifically, such as Figure 1 As shown, the method for predicting source rocks includes:

[0041] S1: Acquire post-stack time-domain seismic data and actual drilling data of encountered hydrocarbon source rocks in the target area;

[0042] Specifically, post-stack time-domain seismic data, acquired through seismic exploration techniques, records the reflection and transmission information of artificially induced seismic waves by underground rock layers, visually showcasing the complexity and diversity of underground geological structures in a time-domain format. Simultaneously, it is also necessary to acquire actual drilling data from wells such as KP11-4-9d and KP11-4-10 that encountered source rocks, confirming the actual presence of source rocks.

[0043] like Figure 2 As shown, S2: Based on actual drilling data and time-domain seismic data, a well-seismic joint calibration is performed to determine the top and bottom interfaces of the source rock in the time-domain seismic data. Based on the top and bottom interfaces, full-area tracking is carried out on the time-domain seismic data to obtain the top and bottom surface data of the source rock.

[0044] The actual drilling data includes sonic data and logging density data. Step S2 includes: obtaining the velocity data and density data of the source rock based on the sonic data and logging data; performing well-seismic joint calibration on the time-domain seismic data based on the velocity data and density data of the source rock to determine the top and bottom interfaces of the time-domain seismic data corresponding to the source rock; and conducting full-area tracking on the time-domain seismic data to obtain the top and bottom surface data of the source rock.

[0045] The wave resistance characteristics of source rocks include three-peak, two-valley and two-peak, two-valley features. Well logging data from actual drilling includes sonic transit time curves and density curves, while time-domain seismic data includes actual seismic records.

[0046] Among them, the joint well-seismic calibration of time-domain seismic data based on the wave impedance characteristics of source rocks to determine the top and bottom interfaces of the time-domain seismic data corresponding to the source rocks includes: using three-dimensional seismic phase tracking technology to track the time-domain seismic data based on the top and bottom interfaces in the actual drilling data to determine the top and bottom interfaces of the time-domain seismic data corresponding to the source rocks.

[0047] Specifically, the wave resistance characteristics of the source rocks (such as "three peaks and two valleys" and "two peaks and two valleys") were extracted from the logging data of actual drilled wells. Then, based on these characteristics, joint well-seismic calibration was performed on the time-domain seismic data. By comparing the cross-correlation (frequency, wave resistance, and stratigraphic matching) between the synthetic seismic record and the actual seismic record, the top and bottom interfaces of the source rocks in the seismic data were accurately determined. The fine calibration of wells KP11-4-9d and KP11-4-10 verified the accuracy of the wave resistance characteristics and revealed the specific morphology of the source rock sections on the seismic profile. Next, a full-area tracing was performed on the time-domain seismic data to obtain complete data on the top and bottom surfaces of the source rocks.

[0048] like Figure 3 and Figure 4As shown, S3: Based on the actual drilling data after well-seismic joint calibration, time-domain seismic data is analyzed using time-spectrum analysis to obtain the characteristic frequencies of source rocks.

[0049] Furthermore, based on the actual drilling data after well-seismic joint calibration, time-domain seismic data is analyzed using time-spectrum analysis to obtain the characteristic frequency range of the source rock. The average value of the characteristic frequency range of the source rock is calculated to obtain the characteristic frequency of the source rock.

[0050] Specifically, based on the actual drilling data after joint well-seismic calibration, time-spectrum analysis technology was used to conduct an in-depth analysis of the time-domain seismic data. The main purpose of this analysis was to reveal the characteristic frequencies of source rocks in the time-domain seismic data. Specifically, by applying time-spectrum analysis to the time-domain seismic data, the frequency range in which the characteristic changes of source rocks are most significant was identified, namely the characteristic frequency band, which was determined to be 22-28Hz.

[0051] To more accurately quantify the characteristic frequencies of the source rock, the frequency values ​​within the characteristic frequency range were calculated, and their average values ​​were obtained, thus yielding the characteristic frequency of the source rock, which is 25 Hz. This result is based on averaging the 22-28 Hz characteristic frequency range. It is worth noting that this 22-28 Hz characteristic frequency range was obtained directly through single-well time-spectrum analysis, reflecting the unique frequency characteristics of the source rock in its seismic response.

[0052] S4: Based on the characteristic frequencies of the source rocks, the time-domain seismic data is interpreted by frequency division to determine the seismic frequency division data corresponding to the characteristic frequencies of the source rocks.

[0053] Furthermore, the time-domain seismic data is subjected to structurally guided filtering. Based on the characteristic frequencies of the source rocks, the time-domain seismic data after structurally guided filtering is interpreted by frequency division to determine the seismic frequency division data corresponding to the characteristic frequencies of the source rocks.

[0054] Specifically, to improve the accuracy and reliability of frequency division interpretation, a structure-guided filtering technique was introduced. Structure-guided filtering is a filtering method based on the strike of geological structures, which can remove noise interference from seismic data while preserving the characteristics of geological structures. Time-domain seismic data is first subjected to structure-guided filtering to eliminate unnecessary noise and improve the signal-to-noise ratio of the seismic data. Subsequently, frequency division interpretation is performed on the filtered seismic data based on the characteristic frequencies of source rocks. This step ensures that frequency division analysis is performed on cleaner and clearer seismic data, thereby improving the accuracy and reliability of the seismic frequency division data corresponding to the characteristic frequencies of source rocks.

[0055] S5: Convert the time-domain seismic frequency-division data into frequency-domain seismic frequency-division data, and use the Shannon interpolation method to perform interpolation processing on the frequency-domain seismic frequency-division data.

[0056] Furthermore, the Fourier transform method is used to convert the time-domain seismic frequency-division data into frequency-domain seismic frequency-division data.

[0057] Specifically, by applying Fourier transform, the data was converted from the time domain to the frequency domain. Then, to optimize the quality of the frequency-domain seismic frequency-division data, the Sinc interpolation method (i.e., the Shannon interpolation algorithm) was introduced. This algorithm uses the sinc function as the interpolation kernel and, through refined convolution operations, effectively eliminates discontinuities and noise in the data, significantly improving the smoothness, resolution, and signal-to-noise ratio. This process not only enhances the data's detail representation but also provides more accurate and reliable data support for subsequent geological structure analysis, source rock distribution assessment, and reservoir characteristic studies.

[0058] like Figure 5 As shown, S6: Based on the top and bottom surface data of the source rock and the frequency domain seismic frequency division data after interpolation, waveform clustering between the top and bottom surfaces is performed to obtain a waveform clustering planar map of the target area reflecting the planar distribution of the source rock.

[0059] Specifically, Figure 5 This figure demonstrates the application of waveform clustering technology based on characteristic frequencies in characterizing the distribution of source rocks. Through waveform clustering analysis, it correlates specific waveform patterns in seismic data with the geological characteristics of source rocks, thereby revealing the distribution of source rocks in the subsurface space.

[0060] exist Figure 5 In the model, Models 1 and 2 exhibit typical "valley-peak" waveform characteristics, which primarily reflect the seismic facies features of source rocks. Seismic facies is a comprehensive representation of the characteristics of subsurface geological bodies in seismic data. The "valley-peak" features identified through waveform clustering technology provide important evidence for identifying source rocks. In contrast, Models 3 and 4 mainly reflect geological structural features such as troughs and faults. These features also manifest as specific waveform patterns in seismic data, but they differ from the "valley-peak" characteristics of source rocks. Through waveform clustering analysis, these different waveform patterns can be distinguished, thereby more accurately identifying source rocks and other geological structures.

[0061] Based on top and bottom surface data of source rocks and frequency-domain seismic data after interpolation, waveform clustering analysis was performed between the top and bottom surfaces. This analysis fully considered the vertical variation characteristics of source rocks and the high-resolution information in the frequency-domain seismic data. Through clustering analysis, waveform clustering plane maps reflecting the planar distribution of source rocks in the target area were obtained. In the waveform clustering plane map, areas with waveform characteristics identical to those in Model 1 and Model 2 were identified as source rocks. These areas are marked with specific colors or symbols on the map, visually demonstrating the planar distribution of source rocks. This result not only provides direct evidence of the distribution of source rocks but also provides important geological basis for subsequent oil and gas exploration and development work.

[0062] The present invention also provides a system for predicting source rocks, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to perform the steps of any of the above-described methods for predicting source rocks.

[0063] The present invention also provides a computer-readable storage medium having a computer program / instructions stored thereon, wherein the computer program / instructions, when executed by a processor, implement the steps of the method for predicting source rocks described above.

[0064] Beneficial effects: By acquiring post-stack time-domain seismic data and actual drilling data of the source rocks encountered in the target area; performing joint well-seismic calibration based on the actual drilling data and time-domain seismic data to determine the top and bottom interfaces of the source rocks in the time-domain seismic data, and conducting full-area tracking based on the top and bottom interfaces in the time-domain seismic data to obtain the top and bottom surface data of the source rocks; using time-spectrum analysis to analyze the time-domain seismic data based on the joint well-seismic calibration to obtain the characteristic frequencies of the source rocks; performing frequency division interpretation of the time-domain seismic data according to the characteristic frequencies of the source rocks to determine the seismic frequency division data corresponding to the characteristic frequencies of the source rocks; converting the time-domain seismic frequency division data into frequency-domain seismic frequency division data, and using the Shannon interpolation method to interpolate the frequency-domain seismic frequency division data to determine the interpolated frequency-domain seismic frequency division data; performing waveform clustering between the top and bottom surfaces based on the source rock top and bottom surface data and the interpolated frequency-domain seismic frequency division data to obtain a waveform clustering planar map reflecting the planar distribution of the source rocks in the target area. This invention precisely locates the position of source rocks in time-domain seismic data through joint well-seismic calibration and full-area tracking. Combined with time-spectrum analysis to extract characteristic frequencies, it effectively reduces waveform distortion caused by formation thickness variations, improving the stability and reliability of waveform clustering. Furthermore, it efficiently performs waveform clustering on interpolated frequency-domain seismic data, intuitively displaying the planar distribution characteristics of source rocks, significantly improving the efficiency and success rate of oil and gas exploration.

[0065] It is understood that the above embodiments only illustrate preferred embodiments of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can freely combine the above technical features without departing from the concept of the present invention, and can also make several modifications and improvements, all of which fall within the protection scope of the present invention. Therefore, all equivalent transformations and modifications made with respect to the scope of the claims of the present invention should fall within the scope of the claims of the present invention.

Claims

1. A method for predicting source rocks of hydrocarbons, characterized in that, The method includes: S1: Acquire time-domain seismic data and actual drilling data of the target area after stacking; S2: Based on the actual drilling data, extract the wave resistance characteristics of the source rock, and then perform well-seismic joint calibration on the time-domain seismic data based on the wave resistance characteristics to determine the top and bottom interfaces of the source rock in the time-domain seismic data. Based on the top and bottom interfaces, conduct full-area tracking on the time-domain seismic data to obtain the top and bottom surface data of the source rock; wherein, the wave resistance characteristics include three-peak and two-valley characteristics and two-peak and two-valley characteristics; S3: Based on the actual drilling data after well-seismic joint calibration, time-domain seismic data is analyzed using time-frequency analysis to obtain the characteristic frequencies of the source rock; S4: Based on the characteristic frequencies of the source rock, the time-domain seismic data is interpreted by frequency division to determine the seismic frequency division data corresponding to the characteristic frequencies of the source rock; S5: Convert the time-domain seismic frequency division data into frequency-domain seismic frequency division data, and use the Shannon interpolation method to interpolate the frequency-domain seismic frequency division data to determine the interpolated frequency-domain seismic frequency division data. S6: Based on the top and bottom surface data of the source rock and the interpolated frequency domain seismic frequency division data, perform waveform clustering between the top and bottom surfaces to obtain a waveform clustering planar map of the target area reflecting the planar distribution of the source rock.

2. The method for predicting source rocks according to claim 1, characterized in that, The actual drilling data includes acoustic data and logging density data, and S2 includes: Based on the acoustic data and the well logging density data, velocity data and density data of the source rock are obtained; Based on the velocity and density data of the source rock, the time-domain seismic data is jointly calibrated by well and seismic data to determine the top and bottom interfaces of the time-domain seismic data corresponding to the source rock. Full-area tracking is then carried out on the time-domain seismic data to obtain the top and bottom surface data of the source rock.

3. The method for predicting source rocks according to claim 2, characterized in that, The actual drilling data includes sonic transit time curves and density curves, and the time-domain seismic data includes actual seismic records. The method of performing well-seismic joint calibration of the time-domain seismic data based on the wave impedance characteristics to determine the top and bottom interfaces of the time-domain seismic data corresponding to the source rock includes: The acoustic transit time curve and the density curve are synthesized to obtain a synthetic seismic record. By comparing the cross-correlation between the actual seismic records and the synthetic seismic records, the top and bottom interfaces of the time-domain seismic data corresponding to the source rocks are determined.

4. The method for predicting source rocks according to claim 3, characterized in that, The process of determining the top and bottom interfaces of the time-domain seismic data corresponding to the source rock by comparing the cross-correlation between actual seismic records and synthetic seismic records includes: Based on the top and bottom interfaces in the actual drilling data, three-dimensional seismic phase tracking technology is used to track the time-domain seismic data to determine the top and bottom interfaces of the time-domain seismic data corresponding to the source rock.

5. The method for predicting source rocks according to claim 1, characterized in that, S3 includes: Based on the actual drilling data after well-seismic joint calibration, time-domain seismic data is analyzed using spectral analysis to obtain the characteristic frequency range of the source rock. The average value of the characteristic frequency range of the source rock is calculated to obtain the characteristic frequency of the source rock.

6. The method for predicting source rocks according to claim 1, characterized in that, S4 includes: The time-domain seismic data is subjected to tectonic-guided filtering. Based on the characteristic frequencies of the source rocks, the time-domain seismic data after tectonic-guided filtering is interpreted by frequency division to determine the seismic frequency division data corresponding to the characteristic frequencies of the source rocks.

7. The method for predicting source rocks according to claim 1, characterized in that, The process of converting the time-domain seismic frequency-divided data into frequency-domain seismic frequency-divided data includes: The time-domain seismic frequency-division data is converted into frequency-domain seismic frequency-division data using the Fourier transform method.

8. A system for predicting source rocks of hydrocarbons, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method for predicting source rocks according to any one of claims 1-7.

9. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method for predicting source rocks according to any one of claims 1-7.

Images