Three-dimensional seismic structure automatic interpretation system based on artificial intelligence
Through the three-dimensional seismic structure automatic interpretation system based on artificial intelligence, using multi-scale decomposition and fusion technology, the problem of being limited to a single scale in traditional seismic data interpretation is solved, and high-precision automatic interpretation of complex seismic structures is achieved, and the explanation is improved. Efficiency and reliability of results.
Patent Information
- Application Number
- CN202411307188.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-19
- Publication Date
- 2024-12-03
AI Technical Summary
Traditional seismic tectonic interpretation methods are limited to a single scale, making it difficult to accurately identify complex geological structures, and manual analysis is time-consuming and labor-intensive, susceptible to human-based subjective factors, so the reliability and consistency of interpretation results are difficult to guarantee.
Adopt a three-dimensional seismic structure automatic interpretation system based on artificial intelligence, including seismic sensing module, sensing correction module, detection scale constraint module, multi-scale decomposition module, multi-scale geological structure detection module and multi-scale interpretation fusion module, through multi-scale decomposition and fusion technology to automatically interpret geological structures and establish high-precision geological structure models.
High-precision automatic interpretation of complex seismic structures is achieved, the interpretation efficiency and reliability of results are improved, the human subjective influence is reduced, and the consistency and accuracy of interpretation results are ensured.
Smart Images

Figure CN119065007A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data processing, and in particular to an automatic interpretation system for three-dimensional seismic structures based on artificial intelligence. Background Art
[0002] In the field of seismic exploration, accurate interpretation of geological structures is of great significance for oil and gas resource exploration, mineral resource detection, and earthquake disaster warning. Traditional seismic structural interpretation usually relies on manual analysis by experts based on two-dimensional or three-dimensional seismic data. However, with the increasing complexity of the exploration area, traditional methods are susceptible to the limitations of a single scale when processing large-scale, multi-dimensional seismic data, making it more difficult to accurately identify and interpret complex geological structures. In addition, the manual analysis process is time-consuming and labor-intensive, and is easily affected by human subjective factors, making it difficult to guarantee the reliability and consistency of the interpretation results. Summary of the invention
[0003] The embodiment of the present application provides a three-dimensional seismic structure automatic interpretation system based on artificial intelligence, which solves the technical problem that the traditional seismic data interpretation process is limited to a single scale and it is difficult to accurately identify complex geological structures.
[0004] In view of the above problems, an embodiment of the present application provides a three-dimensional seismic structure automatic interpretation system based on artificial intelligence.
[0005] The embodiment of the present application provides a 3D seismic structure automatic interpretation system based on artificial intelligence, the system comprising:
[0006] A seismic sensing module, the seismic sensing module is used to obtain a three-dimensional seismic data stream of a target area; a seismic sensing correction module, the seismic sensing correction module is used to perform sensing error correction on the three-dimensional seismic data stream to obtain a three-dimensional seismic data set; a detection scale constraint module, the detection scale constraint module is used to perform geological structure detection scale constraints based on the three-dimensional seismic data set to determine a multi-level geological structure detection scale; a multi-scale decomposition module, the multi-scale decomposition module is used to perform multi-scale decomposition on the three-dimensional seismic data set according to the multi-level geological structure detection scale to obtain a multi-scale seismic data group; a multi-scale geological structure detection module, the multi-scale geological structure detection module is used to perform geological structure detection based on the multi-scale seismic data group to obtain a multi-scale geological structure interpretation result; a multi-scale interpretation fusion module, the multi-scale interpretation fusion module is used to fuse the multi-scale geological structure interpretation results according to the multi-level geological structure detection scale to establish a geological structure interpretation model.
[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0008] The three-dimensional seismic data stream of the target area is obtained through the seismic sensing module. Then, the three-dimensional seismic data stream is corrected for sensing errors through the seismic sensing correction module to obtain a three-dimensional seismic data set. Next, the detection scale constraint module constrains the geological structure detection scale according to the three-dimensional seismic data set and determines the multi-level geological structure detection scale. Next, the multi-scale decomposition module performs multi-scale decomposition on the three-dimensional seismic data set according to the multi-level geological structure detection scale to obtain a multi-scale seismic data group. After that, the multi-scale geological structure detection module performs geological structure detection based on the multi-scale seismic data group, and finally obtains the multi-scale geological structure interpretation result. Finally, the multi-scale interpretation fusion module fuses the multi-scale geological structure interpretation results according to the multi-level geological structure detection scale to establish a geological structure interpretation model. It solves the technical problem that the traditional seismic data interpretation process is limited to a single scale and it is difficult to accurately identify complex geological structures, and achieves the technical effect of high-precision automatic interpretation of complex seismic structures. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0010] Figure 1 A schematic diagram of the structure of a 3D seismic structure automatic interpretation system based on artificial intelligence provided in an embodiment of the present application;
[0011] Figure 2 A schematic diagram of the execution flow of the seismic sensing correction module of the artificial intelligence-based three-dimensional seismic structure automatic interpretation system provided in an embodiment of the present application.
[0012] Explanation of the accompanying drawings: seismic sensing module 11, seismic sensing correction module 12, detection scale constraint module 13, multi-scale decomposition module 14, multi-scale geological structure detection module 15, multi-scale interpretation fusion module 16. DETAILED DESCRIPTION
[0013] The embodiment of the present application solves the technical problem that the traditional seismic data interpretation process is limited to a single scale and has difficulty in accurately identifying complex geological structures by providing a three-dimensional seismic structure automatic interpretation system based on artificial intelligence.
[0014] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0015] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or are inherent to these processes, methods, products or devices.
[0016] Examples, such as Figure 1 As shown, the present application provides a 3D seismic structure automatic interpretation system based on artificial intelligence. The system includes:
[0017] A seismic sensing module 11 is used to obtain a three-dimensional seismic data stream of a target area.
[0018] The seismic sensing module monitors seismic fluctuations in the target area in real time by deploying seismic sensors on the ground, underground or other suitable locations. The seismic data collected by these sensors is a three-dimensional seismic data stream, which is used to reflect the propagation of seismic waves in the target area.
[0019] The seismic sensing correction module 12 is used to perform sensing error correction on the three-dimensional seismic data stream to obtain a three-dimensional seismic data set.
[0020] The seismic sensor correction module is used to correct the initially acquired 3D seismic data stream to ensure the accuracy and consistency of the data. During the seismic data acquisition process, the sensor will be affected by various factors, such as environmental noise, sensor sensitivity differences, interference during data transmission, etc. These factors will introduce errors and cause the original seismic data to be inaccurate. By correcting the 3D seismic data stream and obtaining a 3D seismic data set, the problem of inaccurate geological structure detection results caused by data errors can be avoided, thereby improving the accuracy and reliability of the interpretation.
[0021] Furthermore, if Figure 2 As shown, the seismic sensing correction module 12 is used to perform sensing error correction on the three-dimensional seismic data stream to obtain a three-dimensional seismic data set, including:
[0022] Extract a first seismic sensing parameter from the three-dimensional seismic data stream; retrieve first sensing device status information corresponding to the first seismic sensing parameter according to the seismic sensing module; perform anomaly detection according to the first sensing device status information to obtain a first sensing anomaly detection result; based on the first sensing anomaly detection result, correct the first seismic sensing parameter according to a seismic sensing error correction model to obtain a first seismic sensing optimization parameter; add the first seismic sensing optimization parameter to the three-dimensional seismic data set, and continue to perform error correction on the three-dimensional seismic data stream according to the seismic sensing error correction model to generate the three-dimensional seismic data set.
[0023] Specifically, initial data related to seismic activity, namely, first seismic sensing parameters, are extracted from the three-dimensional seismic data stream. The first seismic sensing parameters include information such as vibration amplitude, frequency, and waveform, which serve as the basic input for sensing error correction. According to the records in the seismic sensing module, first sensing device status information corresponding to the first seismic sensing parameter is retrieved. The first sensing device status information includes factors such as the current working status of the sensor, historical working records, and installation environment, so as to identify abnormal conditions that may occur during equipment operation. The first sensing device status information is used to perform abnormal detection on the equipment operation to identify deviations or faults in the sensing process. For example, if the sensor is disturbed by the external environment (such as temperature fluctuations, equipment aging, etc.) when collecting data, this information will be marked and a first sensor anomaly detection result will be generated; based on the first sensor anomaly detection result, the seismic sensor error correction model is called to correct the detected anomaly and generate the corrected first seismic sensor optimization parameters; the first seismic sensor optimization parameters are added to the three-dimensional seismic data set to replace the original error data. At the same time, the three-dimensional seismic data stream continues to be dynamically corrected for errors to further optimize the overall data and ensure the high accuracy and consistency of the entire data set.
[0024] Furthermore, based on the first sensing anomaly detection result, the first seismic sensing parameter is corrected according to a seismic sensing error correction model to obtain a first seismic sensing optimization parameter, including:
[0025] Perform an earthquake sensing loss evaluation based on the first sensing anomaly detection result to obtain a first earthquake sensing loss coefficient; determine whether the first earthquake sensing loss coefficient is greater than / equal to an earthquake sensing loss threshold; if the first earthquake sensing loss coefficient is greater than / equal to the earthquake sensing loss threshold, activate the earthquake sensing error correction model; based on the first sensing anomaly detection result and the first earthquake sensing parameter, output the first earthquake sensing optimization parameter according to the earthquake sensing error correction model.
[0026] Specifically, based on the first sensor anomaly detection result, the loss in the seismic sensing process is evaluated, and the first seismic sensing loss coefficient can be calculated by comparing the sensing parameters under normal conditions with the sensing parameters under current conditions. The first seismic sensing loss coefficient reflects the degree of degradation of seismic data quality due to the abnormality of the sensing equipment. It is determined whether the first seismic sensing loss coefficient is greater than or equal to a preset seismic sensing loss threshold. The seismic sensing loss threshold is a reference standard, which is preset as the maximum acceptable error range for sensing data. If the loss coefficient is lower than the threshold, it indicates that the sensing data error is within an acceptable range and no further correction is required. If the loss coefficient exceeds or is equal to the threshold, it indicates that the data error is within an acceptable range. The difference is too large and must be corrected; when the system detects that the first seismic sensing loss coefficient is greater than or equal to the seismic sensing loss threshold, the seismic sensing error correction model will be automatically activated. The seismic sensing error correction model is established based on a large amount of experimental data, theoretical analysis and prior knowledge, and can predict and compensate for errors caused by abnormal state of sensing equipment; based on the first sensing anomaly detection result and the original first seismic sensing parameter, the seismic sensing error correction model starts to perform correction operations. The model performs correction calculations by comparing the anomaly detection information with the normal data, adjusts the deviation, and generates the first seismic sensing optimization parameters. The optimized parameters can reflect sensing data that is closer to actual seismic activity, thereby improving the accuracy of the data.
[0027] The detection scale constraint module 13 is used to perform geological structure detection scale constraints according to the three-dimensional seismic data set and determine multi-level geological structure detection scales.
[0028] The detection scale constraint module receives the three-dimensional seismic data set processed by the seismic sensor correction module, analyzes the geological structure information contained therein, and sets multiple detection scales based on the analysis results, combined with factors such as the complexity of the geological structure, depth range, and data resolution. Each scale corresponds to a different geological structure level and detection accuracy. For example, a larger scale can be used to identify macroscopic structural features (such as faults and folds), while a smaller scale is used to detect detailed features (such as cracks and microstructures).
[0029] Furthermore, the detection scale constraint module 13 is used to perform geological structure detection scale constraints according to the three-dimensional seismic data set to determine multi-level geological structure detection scales, including:
[0030] According to the three-dimensional seismic data set, geological risk prediction is performed on the target area to determine the predicted geological risk coefficient; a geological risk coefficient sample set and a geological structure detection scale sample set are loaded; the geological risk coefficient sample set and the geological structure detection scale sample set are sorted to build a geological structure detection scale constraint map; the predicted geological risk coefficient is input into the geological structure detection scale constraint map, and the multi-level geological structure detection scale is output.
[0031] Preferably, geological risk prediction is performed on the target area based on the three-dimensional seismic data set. The expert team can identify and analyze the geological features in the seismic data, such as changes in stratum thickness, fault systems, lithology changes, etc., evaluate the possible geological risk factors in the target area, such as fault activity, crustal movement, etc., and obtain a predicted geological risk coefficient. The predicted geological risk coefficient is used to represent the potential geological risk level of the area. The higher the value, the greater the geological risk; load the pre-stored geological risk coefficient sample set and geological structure detection scale sample set. The geological risk coefficient sample set contains the risk coefficients of multiple known geological risk areas and their corresponding geological features. The geological structure detection scale sample set records the geological structure detection scales used in different geological risk areas and their effect evaluations. Estimate; organize the geological risk factor sample set and the geological structure detection scale sample set, and through statistical analysis and data mining, find the correlation between the geological risk factor and the geological structure detection scale. Based on these correlation rules, build a geological structure detection scale constraint map. The geological structure detection scale constraint map can be regarded as a relationship model, which defines the correlation between the geological risk factor and different detection scales; input the predicted geological risk factor into the geological structure detection scale constraint map, and output the multi-level geological structure detection scale that matches the geological risk factor of the target area according to the mapping relationship of the map. These scales can cover different geological characteristics from macro to micro, ensuring that more refined detection scales are used in areas with higher geological risks to improve the precision and accuracy of detection.
[0032] The multi-scale decomposition module 14 is used to perform multi-scale decomposition on the three-dimensional seismic data set according to the multi-level geological structure detection scale to obtain a multi-scale seismic data set.
[0033] The multi-scale decomposition module decomposes the three-dimensional seismic data set into multiple levels of data groups according to the multi-level geological structure detection scales output by the detection scale constraint module. Each level corresponds to a different detection scale, so as to analyze the geological structure from multiple perspectives from macro to micro. The large scale can capture wide-area geological features, such as faults and folds, while the small scale is used to identify finer structures, such as cracks and tiny deformations.
[0034] During the decomposition process, the system generates seismic data groups at different scales, each of which contains characteristic information of three-dimensional seismic data within the detection scale.
[0035] The multi-scale geological structure detection module 15 is used to perform geological structure detection according to the multi-scale seismic data group to obtain a multi-scale geological structure interpretation result.
[0036] The multi-scale geological structure detection module receives the multi-scale seismic data groups generated by the multi-scale decomposition module. These data groups reflect the seismic data at different detection scales, covering multi-level information from macroscopic geological structures to microscopic structures. The multi-scale geological structure detection module detects geological structures according to the scales corresponding to each data group. For example, the larger scale data group is used to identify regional or large-scale geological features, such as faults, folds, and large rock formations; while the smaller scale data group is used to detect more subtle geological structures, such as small cracks, local deformations, or microstructures; after completing the geological structure detection at each scale, the geological structure interpretation results of each scale, namely the multi-scale geological structure interpretation results, are generated. The multi-scale geological structure interpretation results include detailed information such as the morphology, location, depth, size, etc. of geological structures of different scales in the target area.
[0037] Furthermore, the multi-scale geological structure detection module 15 is used to perform geological structure detection according to the multi-scale seismic data group to obtain a multi-scale geological structure interpretation result, including:
[0038] According to the multi-scale seismic data group, a first-scale seismic data group corresponding to a first geological structure detection scale is extracted; based on the first geological structure detection scale, integrated fusion learning is performed according to multiple geological structure detection learning models to build a first-scale geological structure detection channel; based on the first-scale seismic data group, a first-scale geological structure detection result is obtained according to the first-scale geological structure detection channel; the first-scale geological structure detection result is added to the multi-scale geological structure interpretation result, and geological structure detection is continued on the multi-scale seismic data group to obtain the multi-scale geological structure interpretation result.
[0039] Specifically, according to the multi-scale seismic data group, a first-scale seismic data group corresponding to the first geological structure detection scale is extracted, and the first-scale seismic data group reflects the seismic data characteristics of the target area at a specific scale; based on the first geological structure detection scale, multiple geological structure detection learning models (such as convolutional neural networks, support vector machines, random forests, etc.) are used for integrated fusion learning. Integrated learning can be achieved through a variety of methods, such as voting method, averaging method or stacking method (Stacking) to improve the accuracy and robustness of the detection results. This process builds a first-scale geological structure detection channel, which is used to analyze the first-scale geological structure detection channel. seismic data group; input the first-scale seismic data group into the first-scale geological structure detection channel to obtain the first-scale geological structure detection results, including parameters such as the location, strike, and dip of the fault, as well as information such as the morphology and amplitude of the fold; add the first-scale geological structure detection results to the final multi-scale geological structure interpretation results; continue to use the same process to perform geological structure detection on seismic data groups of other scales. For each scale, a corresponding geological structure detection channel will be established and corresponding detection results will be obtained; by detecting data groups of all scales and integrating the detection results of each scale, a complete multi-scale geological structure interpretation result is finally generated.
[0040] Furthermore, based on the first geological structure detection scale, integrated fusion learning is performed according to multiple geological structure detection learning models to build a first-scale geological structure detection channel, including:
[0041] Based on the first geological structure detection scale, geological structure detection record retrieval is performed to obtain sample scale seismic data records and sample scale geological structure detection result records;
[0042] According to the sample-scale seismic data records and the sample-scale geological structure detection result records, supervised training is performed on the multiple geological structure detection learning models to generate multiple geological structure detectors;
[0043] Using the output data sets of the plurality of geological structure detectors as input information and the sample-scale geological structure detection result records as output information, training a geological structure detection amalgamator;
[0044] Merging the plurality of geological structure detectors as parallel independent nodes to generate a geological structure detection processing layer;
[0045] The geological structure detection processing layer is merged with the input layer of the geological structure detection fuser to generate the first-scale geological structure detection channel.
[0046] Specifically, based on the first geological structure detection scale, the system will retrieve historical geological structure detection records, extract sample-scale seismic data records and corresponding sample-scale geological structure detection result records; use sample-scale seismic data records and sample-scale geological structure detection result records to supervise the training of multiple geological structure detection learning models, and the supervised training optimizes the model through labeled input and output data so that it can learn how to identify geological structures from seismic data, thereby generating multiple trained geological structure detectors, each of which represents an independent detection model that can handle different geological structure detection tasks; use the output data sets of multiple geological structure detectors as input information, and use sample-scale geological structure detection result records as output information for training to generate a geological structure detection fusion device, the geological structure The detection fusion device is used to integrate the results of multiple detectors, enhance the accuracy and robustness of the final detection results, and make the detection results more comprehensive through integrated learning methods; multiple geological structure detectors are merged as parallel independent nodes to generate a geological structure detection processing layer. Each geological structure detector acts as a node to independently process a specific geological structure detection task. The system speeds up the overall detection efficiency through parallel processing and ensures that multiple different detection models work in parallel at the same scale; the generated geological structure detection processing layer is merged with the input layer of the geological structure detection fusion device to form a complete first-scale geological structure detection channel. The first-scale geological structure detection channel can more accurately identify the geological structure characteristics at the first geological structure detection scale by fusing the outputs of multiple parallel detectors and combining the integrated learning ability of the fusion device.
[0047] The multi-scale interpretation fusion module 16 is used to fuse the multi-scale geological structure interpretation results according to the multi-level geological structure detection scales to establish a geological structure interpretation model.
[0048] The multi-scale interpretation fusion module receives the geological structure interpretation results at different detection scales output by the multi-scale geological structure detection module. These interpretation results correspond to different geological structure detection scales, covering geological structure characteristics from macro to micro. The multi-scale interpretation fusion module fuses the interpretation results of each scale according to the multi-level geological structure detection scale to generate a geological structure interpretation model. The geological structure interpretation model integrates geological structure information at different scales and can fully reflect the geological structure characteristics of the target area, including a large range of geological structures and accurate descriptions of detail levels.
[0049] Furthermore, the multi-scale interpretation fusion module 16 is used to fuse the multi-scale geological structure interpretation results according to the multi-level geological structure detection scales to establish a geological structure interpretation model, including:
[0050] Based on the multi-scale geological structure interpretation results, a multi-scale geological structure model is constructed; based on the multi-level geological structure detection scales, the multi-scale geological structure interpretation results are scale-associated to obtain a multi-scale geological structure association result; based on the multi-scale geological structure association result, the multi-scale geological structure model is integrated to obtain the geological structure interpretation model.
[0051] Specifically, based on the multi-scale geological structure interpretation results obtained from the geological structure detection results at different scales, a multi-scale geological structure model is constructed for each detection scale. The multi-scale geological structure model reflects the geological structure characteristics at different scales, such as macroscopic faults and folds, as well as microscopic cracks and local structures; based on the multi-level geological structure detection scale, the interpretation results at each detection scale are scale-correlated to generate multi-scale geological structure correlation results to ensure the consistency between models at different scales. By correlating the geological features in the multi-scale interpretation results, the overlap, interaction and difference between different scales are resolved to ensure that the geological structure models at different scales can be coordinated with each other; based on the multi-scale geological structure correlation results, the geological structure models of different scales are fused. By integrating information from different scales, the geological structure model can not only show the overall geological characteristics at a large scale, but also refine and show the local microstructure; after completing the fusion of the multi-scale geological structure model, a comprehensive geological structure interpretation model is finally generated. The geological structure interpretation model combines the geological structure information at each scale and can comprehensively and accurately show the geological structure characteristics of the target area.
[0052] Furthermore, based on the multi-scale geological structure interpretation results, a multi-scale geological structure model is constructed, including:
[0053] Based on the multi-scale geological structure interpretation results, the multi-scale geological structure model is evaluated for feature consistency to obtain evaluation coefficients of structural models of various scales; it is determined whether the evaluation coefficients of structural models of various scales satisfy structural model evaluation constraints; if the evaluation coefficients of structural models of various scales do not satisfy the structural model evaluation constraints, a structural model optimization instruction is generated; based on the structural model optimization instruction, the multi-scale geological structure model is optimized for feature consistency mapping according to the multi-scale geological structure interpretation results.
[0054] Specifically, based on the results of multi-scale geological structure interpretation, the characteristic consistency of the geological structure model at each scale is evaluated. By analyzing the similarities, differences and consistency between the geological structure characteristics of different scales (such as faults, cracks, folds, etc.), the structural model evaluation coefficient of each scale is calculated. The structural model evaluation coefficient is used to measure the degree of coordination and consistency between models of different scales; according to the preset structural model evaluation constraints, the structural model evaluation coefficient of each scale is judged. If the evaluation coefficient of one or more scales does not reach the set threshold, it means that the characteristic consistency between these models is insufficient and needs further optimization; if the evaluation coefficient meets the constraints, the consistency of the model has met the expected requirements and the model can be directly used in subsequent fusion step; when the system determines that the evaluation coefficients of some structural models do not meet the constraints, it will generate structural model optimization instructions. The structural model optimization instructions are intended to guide the system to optimize the characteristics of the multi-scale model to ensure that the geological structural characteristics at each scale have higher consistency and reduce conflicts or information loss between models of different scales; based on the structural model optimization instructions, the multi-scale geological structural model is adjusted through feature consistent mapping optimization, and the structural characteristics at each scale can be remapped to make the characteristics between the models more consistent in terms of spatial position, structural morphology, etc. For example, if a structural feature in the large-scale model deviates in the small-scale model, the system will adjust the characteristics of the small-scale model to make it consistent with the large-scale model.
[0055] In summary, the embodiments of the present application have at least the following technical effects:
[0056] The three-dimensional seismic data stream of the target area is obtained through the seismic sensing module. Then, the three-dimensional seismic data stream is corrected for sensing errors through the seismic sensing correction module to obtain a three-dimensional seismic data set. Next, the detection scale constraint module constrains the geological structure detection scale according to the three-dimensional seismic data set and determines the multi-level geological structure detection scale. Next, the multi-scale decomposition module performs multi-scale decomposition on the three-dimensional seismic data set according to the multi-level geological structure detection scale to obtain a multi-scale seismic data group. After that, the multi-scale geological structure detection module performs geological structure detection based on the multi-scale seismic data group, and finally obtains the multi-scale geological structure interpretation result. Finally, the multi-scale interpretation fusion module fuses the multi-scale geological structure interpretation results according to the multi-level geological structure detection scale to establish a geological structure interpretation model. It solves the technical problem that the traditional seismic data interpretation process is limited to a single scale and it is difficult to accurately identify complex geological structures, and achieves the technical effect of high-precision automatic interpretation of complex seismic structures.
[0057] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0058] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.
[0059] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.
Claims
1. The 3D seismic structure automatic interpretation system based on artificial intelligence is characterized by: The system comprises: A seismic sensing module, wherein the seismic sensing module is used to obtain a three-dimensional seismic data stream of a target area; A seismic sensing correction module, the seismic sensing correction module is used to perform sensing error correction on the three-dimensional seismic data stream to obtain a three-dimensional seismic data set; A detection scale constraint module, the detection scale constraint module is used to perform geological structure detection scale constraints according to the three-dimensional seismic data set and determine a multi-level geological structure detection scale; A multi-scale decomposition module, the multi-scale decomposition module is used to perform multi-scale decomposition on the three-dimensional seismic data set according to the multi-level geological structure detection scale to obtain a multi-scale seismic data set; A multi-scale geological structure detection module, which is used to perform geological structure detection according to the multi-scale seismic data group to obtain a multi-scale geological structure interpretation result; A multi-scale interpretation fusion module is used to fuse the multi-scale geological structure interpretation results according to the multi-level geological structure detection scales to establish a geological structure interpretation model.
2. The three-dimensional seismic structure automatic interpretation system based on artificial intelligence according to claim 1, characterized in that: The seismic sensing correction module is used to perform sensing error correction on the three-dimensional seismic data stream to obtain a three-dimensional seismic data set, including: extracting a first seismic sensing parameter within the three-dimensional seismic data stream; According to the seismic sensing module, retrieving first sensing device status information corresponding to the first seismic sensing parameter; Perform anomaly detection according to the first sensor device state information to obtain a first sensor anomaly detection result; Based on the first sensing anomaly detection result, the first seismic sensing parameter is corrected according to a seismic sensing error correction model to obtain a first seismic sensing optimization parameter; The first seismic sensing optimization parameter is added to the three-dimensional seismic data set, and the three-dimensional seismic data stream is continuously error corrected according to the seismic sensing error correction model to generate the three-dimensional seismic data set.
3. The three-dimensional seismic structure automatic interpretation system based on artificial intelligence according to claim 2, characterized in that: Based on the first sensing anomaly detection result, the first seismic sensing parameter is corrected according to a seismic sensing error correction model to obtain a first seismic sensing optimization parameter, including: Performing an earthquake sensing loss evaluation according to the first sensing anomaly detection result to obtain a first earthquake sensing loss coefficient; Determining whether the first earthquake sensing loss coefficient is greater than / equal to an earthquake sensing loss threshold; If the first seismic sensing loss coefficient is greater than / equal to the seismic sensing loss threshold, activating the seismic sensing error correction model; Based on the first sensing anomaly detection result and the first seismic sensing parameter, the first seismic sensing optimization parameter is output according to the seismic sensing error correction model.
4. The three-dimensional seismic structure automatic interpretation system based on artificial intelligence according to claim 1, characterized in that: The detection scale constraint module is used to perform geological structure detection scale constraints according to the three-dimensional seismic data set to determine the multi-level geological structure detection scale, including: Performing geological risk prediction on the target area according to the three-dimensional seismic data set to determine a predicted geological risk coefficient; Loading geological risk factor sample set and geological structure detection scale sample set; Arrange the geological risk factor sample set and the geological structure detection scale sample set, and build a geological structure detection scale constraint map; The predicted geological risk coefficient is input into the geological structure detection scale constraint map, and the multi-level geological structure detection scale is output.
5. The three-dimensional seismic structure automatic interpretation system based on artificial intelligence according to claim 1, characterized in that: The multi-scale geological structure detection module is used to perform geological structure detection according to the multi-scale seismic data group to obtain a multi-scale geological structure interpretation result, including: Extracting a first-scale seismic data group corresponding to a first geological structure detection scale according to the multi-scale seismic data group; Based on the first geological structure detection scale, integrated fusion learning is performed according to multiple geological structure detection learning models to build a first-scale geological structure detection channel; Based on the first-scale seismic data group, and according to the first-scale geological structure detection channel, obtaining a first-scale geological structure detection result; The first-scale geological structure detection result is added to the multi-scale geological structure interpretation result, and the geological structure detection is continued on the multi-scale seismic data group to obtain the multi-scale geological structure interpretation result.
6. The three-dimensional seismic structure automatic interpretation system based on artificial intelligence according to claim 5, characterized in that: Based on the first geological structure detection scale, integrated fusion learning is performed according to multiple geological structure detection learning models to build a first-scale geological structure detection channel, including: Based on the first geological structure detection scale, geological structure detection record retrieval is performed to obtain sample scale seismic data records and sample scale geological structure detection result records; According to the sample-scale seismic data records and the sample-scale geological structure detection result records, supervised training is performed on the multiple geological structure detection learning models to generate multiple geological structure detectors; Using the output data sets of the plurality of geological structure detectors as input information and the sample-scale geological structure detection result records as output information, training a geological structure detection amalgamator; Merging the plurality of geological structure detectors as parallel independent nodes to generate a geological structure detection processing layer; The geological structure detection processing layer is merged with the input layer of the geological structure detection fuser to generate the first-scale geological structure detection channel.
7. The three-dimensional seismic structure automatic interpretation system based on artificial intelligence according to claim 1, characterized in that: The multi-scale interpretation fusion module is used to fuse the multi-scale geological structure interpretation results according to the multi-level geological structure detection scales to establish a geological structure interpretation model, including: Based on the multi-scale geological structure interpretation results, construct a multi-scale geological structure model; Performing scale association on the multi-scale geological structure interpretation results based on the multi-level geological structure detection scales to obtain a multi-scale geological structure association result; Based on the multi-scale geological structure association results, the multi-scale geological structure model is integrated to obtain the geological structure interpretation model.
8. The three-dimensional seismic structure automatic interpretation system based on artificial intelligence according to claim 7, characterized in that: Based on the multi-scale geological structure interpretation results, a multi-scale geological structure model is constructed, including: Based on the multi-scale geological structure interpretation results, the multi-scale geological structure model is evaluated for feature consistency to obtain evaluation coefficients of structural models at each scale; Determining whether the evaluation coefficients of the construction model at each scale meet the construction model evaluation constraints; If the evaluation coefficients of the construction models at each scale do not satisfy the construction model evaluation constraints, a construction model optimization instruction is generated. Based on the structural model optimization instruction, the multi-scale geological structure model is optimized for feature consistent mapping according to the multi-scale geological structure interpretation result.
Citation Information
Cited By
Image processing method and system based on artificial intelligence
CN120294825A
