VSP data acquisition system based on ai big-data model, and processing method therefor
Patent Information
- Application Number
- PCT/CN2026/077409
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-03-11
- Filing Date
- 2026-02-06
- Publication Date
- 2026-09-17
Smart Images
Figure CN2026077409_17092026_PF_FP_ABST
Abstract
Description
VSP Data Acquisition System and Processing Method Based on AI Big Data Model Technical Field
[0001] This invention pertains to the application of AI artificial intelligence big data models in the field of geophysical exploration technology, specifically involving a microseismic monitoring system and processing method based on an AI artificial intelligence big data model. Background Technology
[0002] Artificial Intelligence (AI) is a crucial driving force behind the new round of technological revolution and industrial transformation. It is a new technological science that studies and develops theories, methods, technologies, and application systems to simulate, extend, and expand human intelligence. AI is an interdisciplinary and emerging field based on computer science, integrating computer science, psychology, philosophy, and other disciplines. It aims to understand the essence of intelligence and produce new intelligent machines that can react in a way similar to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems. AI is an important component of the discipline of intelligence; it attempts to understand the essence of intelligence and produce new intelligent machines that can react in a way similar to human intelligence. AI is a very broad science, encompassing robotics, speech recognition, image recognition, natural language processing, expert systems, machine learning, and computer vision.
[0003] Large models refer to machine learning models with a massive number of parameters and complex computational structures. These models are typically built from deep neural networks and have billions or even hundreds of billions of parameters. The purpose of large models is to improve their expressive power and predictive performance, enabling them to handle more complex tasks and data. Large models have wide applications in various fields, including natural language processing, computer vision, speech recognition, and recommender systems. By training on massive amounts of data to learn complex patterns and features, large models possess stronger generalization capabilities and can make accurate predictions on unseen data.
[0004] Big data analytics refers to the process of processing, analyzing, and mining large volumes of high-speed, multi-source, and multi-type data to discover valuable information and knowledge. Core technologies of big data analytics include data storage, data processing, data mining, data analysis, and data visualization. With the development of technologies such as the Internet, artificial intelligence, and the Internet of Things, the scale and complexity of data are constantly increasing, making big data analytics an indispensable part of enterprises and organizations.
[0005] In the field of deep learning, model compression and deployment are important research topics, and model distillation is one effective method. Model distillation was first proposed by Hinton et al. in 2015. Its core idea is to transfer knowledge from a complex large model (teacher model) to a relatively simple small model (student model) through knowledge transfer. Simply put, it uses the prediction probability distribution of the teacher model as a soft label to train the student model, thereby greatly reducing the complexity and computational resource requirements of the model while maintaining high prediction performance, achieving lightweight and efficient modeling.
[0006] A vertical seismic profile (VSP) is a seismic observation method. A vertical profile is the counterpart to a surface seismic profile. This method involves observing the seismic wave field in a well, placing seismic detectors at different depths within the well to record the seismic signals generated by a surface source. The vertical seismic profile (VSP) corresponds to a typical surface-observed seismic profile. Surface-observed seismic profiles generate seismic waves at points near the surface and observe them at detectors arranged along the surface survey line; vertical seismic profiles also generate seismic waves at points near the surface, but they are observed at detectors arranged at different depths along the wellhead. The former uses detectors at the surface and survey lines along the surface, hence it is also called a horizontal (or surface) seismic profile; the latter uses detectors in a well and survey lines arranged vertically along the wellbore, hence it is called a vertical seismic profile. In horizontal seismic profiles, because the geophones are placed on the surface, only upward waves from underground can be received, in addition to direct waves and surface waves propagating along the surface. In vertical seismic profiles, because the geophones are placed inside the formation through a well, both upward and downward propagating waves can be received. This is perhaps the most important characteristic of vertical seismic profiles compared to horizontal seismic profiles. Vertical seismic profiles are actually a type of well-logging method, representing a transformation and development of the already widely used seismic logging (also known as velocity verification blasting) method. The differences between seismic well logging and vertical seismic profiling are as follows: the former utilizes only the recorded first arrival wave, while the latter utilizes both the first and subsequent arrival waves; the former typically has a larger observation point spacing, while the latter has a very small one; the former only utilizes a zero-offset observation system near the wellhead, while the latter also utilizes offset observation systems and multi-offset observation systems with the source deviating from the wellhead; the former primarily aims to determine wave velocity, while the latter mainly studies the stratigraphic profile near the well and investigates the formation and propagation of waves in actual geological media. Furthermore, vertical seismic profiling has, in its development, developed specialized instrument systems, tested a complete set of fieldwork methods, and developed a theoretical foundation for interpretation. Therefore, it has far exceeded the original scope of seismic well logging and has evolved into a complete, independent, and new observation method. Summary of the Invention
[0007] This invention proposes a vertical seismic profile (VSP) data acquisition system and processing method based on an AI artificial intelligence big data model, including a seismic source deployed on the ground or sea surface, a VSP data acquisition sensor array deployed underground, a VSP data acquisition and processing computer workstation system installed at the ground wellhead or offshore platform, a distributed fiber optic acoustic wave sensor DAS modulation and demodulation instrument installed at the ground wellhead or offshore platform, an AI-trained VSP data processing artificial big data model, and a VSP data AI processing model extracted from the AI-trained VSP data processing artificial big data model.
[0008] The seismic source can be one of the following: explosive seismic source, controllable seismic source, hammer seismic source, gas explosion seismic source, electric energy seismic source, electric spark seismic source, air gun seismic source, or plasma seismic source.
[0009] The VSP data acquisition sensor array can be one of the following: a downhole three-component velocity detector, a three-component piezoelectric detector, a three-component acceleration detector, or a three-component fiber optic detector. It can also be one of the following: armored straight or spiral optical cables fixed or wound inside or outside the casing or inside or outside the tubing.
[0010] The seismic source is deployed according to the location requirements of the construction design. It can be excited near the wellhead (zero offset), or far from the wellhead (non-zero offset). It can be moved gradually from the wellhead to both ends (variable offset - Walkaway) to be excited sequentially. It can be moved around the wellhead at different radius distances (Walkaround) to be excited sequentially. It can be uniformly deployed around the wellhead in a three-dimensional manner (three-dimensional VSP) according to a grid pattern to be excited sequentially. It can also be used on the sea surface with an air gun seismic source or a plasma seismic source moving in a circle around the drilling platform as the center to be excited sequentially, or uniformly deployed around the drilling platform in a three-dimensional manner (three-dimensional VSP) according to a grid pattern to be excited sequentially.
[0011] The three-component velocity detector, three-component piezoelectric detector, three-component acceleration detector, and three-component fiber optic detector of the VSP data acquisition sensor array are connected to the VSP data acquisition and processing computer workstation system via cables or optical fiber composite cables.
[0012] The optical cable is connected to the distributed fiber optic acoustic wave sensing DAS modulation and demodulation instrument near the wellhead on the ground via an armored optical cable.
[0013] Using AI-based subsurface 3D velocity modeling software, input the 3D geological model, 3D seismic P-wave velocity model, 3D seismic S-wave velocity model, all sonic logging data, VSP data from other wells, rock physics measurement data, anisotropy coefficient and attenuation coefficient Q value of the VSP data acquisition area, and establish the 3D viscoelastic medium anisotropic P-wave velocity model and S-wave velocity model in the subsurface of the VSP data acquisition area.
[0014] The VSP data processing artificial big data model employs various underground three-dimensional seismic P-wave and S-wave velocity models, including uniform and non-uniform isotropic seismic P-wave and S-wave velocity models, uniform and non-uniform anisotropic seismic P-wave and S-wave velocity models, acoustic seismic velocity models, elastic wave seismic P-wave and S-wave velocity models, viscoelastic medium seismic P-wave and S-wave velocity models, and complex geological structure seismic P-wave and S-wave velocity models. Based on the deployment of the VSP data acquisition sensor array and seismic source, it performs forward modeling to simulate the full-wavefield artificially synthesized VSP seismic records recorded in wells for any underground geological structure, as well as the actual VSP data recorded by various existing geophones over the years. Then, it performs AI training for the VSP data processing artificial big data model.
[0015] The VSP data AI processing model is a VSP data AI processing model obtained by distillation based on the VSP data processing artificial big data model. It is used to process the collected VSP data on-site using a VSP data acquisition and processing computer workstation system on a ground or offshore platform.
[0016] The data acquisition and processing method of the VSP data acquisition system based on the AI artificial intelligence big data model includes the following steps:
[0017] (1) Deploy the optimal VSP data acquisition sensor array in the VSP data acquisition work area according to the construction design and downhole conditions;
[0018] (2) Collect three-dimensional geological models, three-dimensional velocity models, sonic logging data of all wells, VSP data of other wells, rock physical measurement data, anisotropy coefficient and attenuation coefficient Q value in the VSP data acquisition area, and use AI-based underground three-dimensional velocity modeling software to establish three-dimensional viscoelastic medium anisotropic longitudinal wave velocity model and three-dimensional viscoelastic medium anisotropic transverse wave velocity model in the VSP data acquisition area.
[0019] (3) Start the VSP data acquisition and processing computer workstation system or the distributed fiber optic acoustic wave sensor DAS modulation and demodulation instrument.
[0020] (4) In the VSP data acquisition work area, start one (single) or multiple (multiple) seismic sources according to the construction design to excite at each pre-designed seismic source point. At the same time, collect various VSP data together through the VSP data acquisition and processing computer workstation system or the distributed fiber optic acoustic wave sensor DAS modulation and demodulation instrument and the downhole VSP data acquisition sensor array.
[0021] (5) Simultaneously start the VSP data acquisition and processing computer workstation system, which is equipped with the AI real-time VSP data processing model evaporating from the AI-trained VSP data processing artificial big data model;
[0022] (6) Process the VSP data collected in the VSP data acquisition area on-site in the VSP data acquisition and processing computer workstation system;
[0023] (7) Use the AI-based VSP data preprocessing software installed in the VSP data acquisition and processing computer workstation system to perform amplitude-preserving noise reduction, surface static correction, surface or downhole consistency processing, deconvolution processing, and wavefield separation processing of upper and lower shaped waves on the VSP data in step (6).
[0024] (8) Input the VSP data processed in step (7), the three-dimensional viscoelastic medium anisotropic longitudinal wave velocity model, and the three-dimensional viscoelastic medium anisotropic transverse wave velocity model into the distilled VSP data AI processing model.
[0025] (9) The VSP data AI processing model in the VSP data acquisition and processing computer workstation system obtains the amplitude-preserving high-resolution VSP longitudinal wave imaging data volume and the amplitude-preserving high-resolution VSP transverse wave imaging data volume based on the VSP data processed in step (7) and the three-dimensional viscoelastic medium anisotropic longitudinal wave velocity model and the three-dimensional viscoelastic medium anisotropic transverse wave velocity model established in step (2).
[0026] (10) Extract various VSP longitudinal wave data attributes and various VSP shear wave data attributes from the amplitude-preserving high-resolution VSP longitudinal wave imaging data volume and VSP shear wave imaging data volume obtained in step (9).
[0027] (11) Using the multiple VSP longitudinal wave data attributes and multiple VSP transverse wave data attributes extracted in step (10), based on the single attribute data or several combined attribute data that are most sensitive to the comprehensive interpretation target or reservoir physical parameters, a three-dimensional characterization and fine depiction of the geological structure or oil and gas reservoir spatial distribution within a certain range around the VSP data acquisition well is carried out. The distribution range of oil and gas resources in the oil and gas reservoir around the well is tracked, the oil / water boundary or gas / water boundary or oil / gas boundary is determined, and the oil and gas saturation in the oil and gas reservoir is calculated. This enables accurate evaluation of oil and gas resources around underground wells based on AI artificial intelligence big data models and VSP data, providing reliable technical support for the efficient development and production of oil and gas resources and the improvement of oil and gas recovery rate. Attached Figure Description
[0028] Figure 1 is a schematic diagram of the VSP data acquisition system and processing method based on the AI artificial intelligence big data model of the present invention.
[0029] Figure 2 is a schematic diagram of the VSP data acquisition system of the downhole three-component geophone with a ring-shaped seismic source on the ground or sea surface according to the present invention.
[0030] Figure 3 is a schematic diagram of the downhole three-component geophone VSP data acquisition system with ground or sea surface grid-like seismic sources according to the present invention.
[0031] Figure 4 is a schematic diagram of the VSP data acquisition system of armored straight optical cable outside or inside the casing of the well with seismic sources arranged in a grid pattern on the ground or sea surface according to the present invention.
[0032] Figure 5 is a schematic diagram of the VSP data acquisition system of the downhole casing spiral armored optical cable with ground or sea surface grid-like seismic source deployment according to the present invention.
[0033] The attached diagram shows the markings and corresponding component names:
[0034] 1-Earthquake source, 2-VSP data acquisition sensor array, 3-VSP data acquisition and processing computer workstation system, 4-Distributed fiber optic acoustic wave sensing DAS modulation and demodulation instrument, 5-VSP data processing artificial big data model, 6-VSP data AI processing model, 7-Optical cable, 8-Three-dimensional viscoelastic medium anisotropic longitudinal wave velocity model and transverse wave velocity model. Detailed Implementation
[0035] To facilitate understanding of the present invention, a more detailed description is provided below with reference to the accompanying drawings and specific embodiments. Preferred embodiments of the invention are shown in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described in this specification. Rather, these embodiments are provided to provide a more thorough and complete understanding of the disclosure of the present invention. They are not intended to limit the invention, but are merely illustrative, and the advantages of the invention will become clearer and easier to understand by illustrating them.
[0036] Figure 1 shows a schematic diagram of the VSP data acquisition system and processing method based on an AI artificial intelligence big data model according to the present invention. The vertical seismic profile (VSP) data acquisition system based on an AI artificial intelligence big data model of the present invention includes a seismic source 1 deployed on the ground or sea surface, a VSP data acquisition sensor array 2 deployed underground, a VSP data acquisition and processing computer workstation system 3 installed at the wellhead on the ground or on a marine platform, a distributed fiber optic acoustic wave sensing DAS modulation and demodulation instrument 4 installed at the wellhead on the ground or on the marine platform, an AI-trained VSP data processing artificial big data model 5, and a VSP data AI processing model 6 extracted from the VSP data processing artificial big data model 5.
[0037] The seismic source 1 can be one of the following: explosive seismic source, controllable seismic source, hammer seismic source, gas explosion seismic source, electric energy seismic source, electric spark seismic source, air gun seismic source, and plasma seismic source; the VSP data acquisition sensor array 2 can be one of the following: downhole three-component velocity detector, three-component piezoelectric detector, three-component acceleration detector, and three-component fiber optic detector, or it can be one of the following: armored straight or spiral optical cables 7 fixed or wound inside or outside the casing or inside or outside the tubing.
[0038] Figure 2 is a schematic diagram of the downhole three-component geophone VSP data acquisition system with a ring-shaped arrangement of seismic sources on the ground or sea surface according to the present invention. Figure 3 is a schematic diagram of the downhole three-component geophone VSP data acquisition system with a grid-shaped arrangement of seismic sources on the ground or sea surface according to the present invention. The seismic source 1 is arranged according to the location requirements of the construction design. It can be excited near the wellhead (zero offset), excited far from the wellhead (non-zero offset), excited sequentially by gradually moving from the wellhead to both ends (variable offset - walkaway), excited sequentially by moving around the wellhead at different radius distances (walkaround) (Figure 2), excited sequentially by uniformly arranging the seismic sources in a three-dimensional manner (three-dimensional VSP) around the wellhead in a grid pattern (Figure 3), and excited sequentially by moving the seismic source in a circle around the drilling platform on the sea surface using an air gun seismic source or a plasma seismic source (Figure 2) or excited sequentially by uniformly arranging the seismic sources in a three-dimensional manner (three-dimensional VSP) around the drilling platform in a grid pattern (Figure 3).
[0039] Figure 4 is a schematic diagram of the VSP data acquisition system of the downhole casing with armored straight optical cable outside or inside the casing, where the seismic source is arranged in a grid pattern on the ground or sea surface according to the present invention. Figure 5 is a schematic diagram of the VSP data acquisition system of the downhole casing with spiral armored optical cable outside the casing, where the seismic source is arranged in a grid pattern on the ground or sea surface according to the present invention.
[0040] The three-component velocity detector, three-component piezoelectric detector, three-component acceleration detector, and three-component fiber optic detector of the VSP data acquisition sensor array 2 are connected to the VSP data acquisition and processing computer workstation system 3 via cables or optoelectronic composite cables; the optical cable 7 is a downhole fiber optic detector array or an armored straight cable (Figure 4) or an armored spiral cable (Figure 5), and the optical cable 7 is connected to the distributed fiber optic acoustic wave sensing DAS modulation and demodulation instrument 4 near the wellhead on the surface via an armored optical cable.
[0041] Using AI-based subsurface 3D velocity modeling software, the following data were input: 3D geological model, 3D seismic P-wave velocity model, 3D seismic S-wave velocity model, all sonic logging data, VSP data from other wells, rock physics measurement data, anisotropy coefficient, and attenuation coefficient Q value. A 3D viscoelastic medium anisotropic P-wave velocity model and a 3D viscoelastic medium anisotropic S-wave velocity model were established in the subsurface of the VSP data acquisition area.
[0042] The VSP data processing artificial big data model 5 uses various underground three-dimensional seismic velocity models, including uniform and non-uniform isotropic seismic velocity models, uniform and non-uniform anisotropic seismic velocity models, acoustic seismic velocity models, elastic wave seismic velocity models, viscoelastic medium seismic velocity models, and complex geological structure seismic velocity models. Based on the layout of the VSP data acquisition sensor array 2 and the seismic source 1, it performs forward modeling to simulate the full-wave field artificially synthesized VSP seismic records recorded in the well by any underground geological structure, as well as the actual VSP data recorded by various geophones over many years. Then, it performs AI training for the VSP data processing artificial big data model 5.
[0043] The VSP data AI processing model 6 is obtained by distillation based on the VSP data processing artificial big data model 5. It is used to process the collected VSP data on the construction site using the VSP data acquisition and processing computer workstation system 3 on the ground or offshore platform.
[0044] The data acquisition and processing method of the VSP data acquisition system based on the AI artificial intelligence big data model includes the following steps:
[0045] (a) Deploy the optimal VSP data acquisition sensor array 2 in the VSP data acquisition work area according to the construction design and downhole conditions;
[0046] (b) Collect three-dimensional geological models, three-dimensional velocity models, sonic logging data of all wells, VSP data of other wells, rock physical measurement data, anisotropy coefficient and attenuation coefficient Q value in the VSP data acquisition area, and use AI-based underground three-dimensional velocity modeling software to establish three-dimensional viscoelastic medium anisotropic longitudinal wave velocity model and three-dimensional viscoelastic medium anisotropic transverse wave velocity model under the VSP data acquisition area.
[0047] (c) Start the VSP data acquisition and processing computer workstation system 3 or the distributed fiber optic acoustic wave sensor DAS modulation and demodulation instrument 4.
[0048] (d) In the VSP data acquisition work area, one or more seismic sources 1 are activated at each pre-designed seismic source point according to the construction design. At the same time, various VSP data are collected together by the VSP data acquisition and processing computer workstation system 3 or the distributed fiber optic acoustic wave sensor DAS modulation and demodulation instrument 4 and the downhole VSP data acquisition sensor array 2.
[0049] (e) Simultaneously start the VSP data acquisition and processing computer workstation system 3, which is equipped with the AI real-time VSP data processing model 6 evaporating from the AI-trained VSP data processing artificial big data model 5;
[0050] (f) The VSP data collected in the VSP data acquisition work area is processed on-site in the VSP data acquisition and processing computer workstation system 3;
[0051] (g) The VSP data in step (f) is processed by the AI-based VSP data preprocessing software installed in the VSP data acquisition and processing computer workstation system 3 to perform amplitude-preserving noise reduction, surface static correction, surface or downhole consistency processing, deconvolution processing, and AI artificial intelligence wavefield separation processing of upper and lower shaped waves.
[0052] (h) Input the VSP data processed in step (g) and the three-dimensional viscoelastic medium anisotropic longitudinal wave velocity model and the three-dimensional viscoelastic medium anisotropic transverse wave velocity model (8) into the distilled VSP data AI processing model 6;
[0053] (i) The VSP data AI processing model 6 in the VSP data acquisition and processing computer workstation system 3 obtains the amplitude-preserving high-resolution VSP longitudinal wave imaging data volume and the amplitude-preserving high-resolution VSP transverse wave imaging data volume based on the VSP data processed in step (g) and combined with the three-dimensional viscoelastic medium anisotropic longitudinal wave velocity model and the three-dimensional viscoelastic medium anisotropic transverse wave velocity model 8 established in step (b).
[0054] (j) Extract various VSP longitudinal wave data attributes and various VSP shear wave data attributes from the amplitude-preserving high-resolution VSP longitudinal wave imaging data volume and the amplitude-preserving high-resolution VSP shear wave imaging data volume obtained in step (i).
[0055] (k) Utilizing the various VSP P-wave and VSP S-wave data attributes extracted in step (j), based on the single attribute data or several combined attribute data that are most sensitive to the comprehensive interpretation target or reservoir physical parameters, a three-dimensional characterization and detailed delineation of the geological structure or oil and gas reservoir spatial distribution within a certain range around the VSP data acquisition well is performed. This includes tracking the distribution range of oil and gas resources within the oil and gas reservoir around the well, determining the oil / water boundary, gas / water boundary, or oil / gas boundary, and calculating the oil and gas saturation within the oil and gas reservoir. This enables accurate evaluation of oil and gas resources around underground wells based on AI artificial intelligence big data models and VSP data, providing reliable technical support for the efficient development and production of oil and gas resources and improving the recovery rate of oil and gas resources.
Claims
1. A VSP data acquisition system based on an AI artificial intelligence big data model, characterized in that, Including seismic sources deployed on the ground or sea surface (1), and VSP data acquisition sensor arrays deployed underground (2); The VSP data acquisition sensor array (2) is one of the following: downhole three-component velocity detector, three-component piezoelectric detector, three-component acceleration detector, or three-component fiber optic detector. The VSP data acquisition sensor array (2) is connected to the VSP data acquisition and processing computer workstation system (3) via a cable or optical fiber composite cable. The VSP data acquisition and processing computer workstation system (3) is equipped with an AI-trained VSP data processing artificial big data model (5) and a VSP data AI processing model (6) evaporating from the VSP data processing artificial big data model (5). Using AI-based underground three-dimensional velocity modeling software, the three-dimensional geological model, three-dimensional seismic P-wave velocity model, three-dimensional seismic S-wave velocity model, all acoustic logging data, VSP data from other wells, rock physics measurement data, anisotropy coefficient and attenuation coefficient Q value are input to establish a three-dimensional viscoelastic medium anisotropic P-wave velocity model and S-wave velocity model (8) underground in the VSP data acquisition area. Alternatively, the VSP data acquisition sensor array (2) is one of the optical cables (7); the optical cable (7) is a straight or spiral armored optical cable fixed or wound inside or outside the casing or inside or outside the pipe, and the optical cable (7) is connected to the distributed optical fiber acoustic wave sensing DAS modulation and demodulation instrument (4) near the wellhead on the ground through an armored optical cable. The VSP data acquisition and processing computer workstation system (3) or the distributed fiber optic acoustic wave sensing DAS modulation and demodulation instrument (4) is set up near the ground wellhead or on an offshore platform.
2. The VSP data acquisition system based on an AI artificial intelligence big data model according to claim 1, characterized in that, The seismic source (1) is one of the following: explosive seismic source, controllable seismic source, hammer seismic source, gas explosion seismic source, electric energy seismic source, electric spark seismic source, air gun seismic source, or plasma seismic source.
3. The VSP data acquisition system based on an AI artificial intelligence big data model according to claim 1, characterized in that, The seismic source (1) is deployed according to the location requirements of the construction design. It can be excited near the wellhead, i.e., zero offset VSP, or far from the wellhead, i.e., non-zero offset VSP, or gradually moved from the wellhead to both ends to excite sequentially, i.e., Walkaway VSP or variable offset VSP, or moved around the wellhead at different radius distances to excite sequentially, i.e., Walkaround VSP, or uniformly deployed in a three-dimensional manner around the wellhead in a grid manner to excite sequentially, i.e., land-based three-dimensional VSP, or on the sea surface, a gas gun source or plasma source is used to move in a circle around the drilling platform as the center to excite sequentially, or uniformly deployed in a three-dimensional manner around the drilling platform in a grid manner to excite sequentially, i.e., offshore three-dimensional VSP.
4. The VSP data acquisition system based on an AI artificial intelligence big data model according to claim 1, characterized in that, The VSP data processing artificial big data model (5) adopts various underground three-dimensional seismic P-wave and S-wave velocity models, including uniform and non-uniform isotropic seismic P-wave and S-wave velocity models, uniform and non-uniform anisotropic seismic P-wave and S-wave velocity models, acoustic seismic velocity models, elastic wave seismic P-wave and S-wave velocity models, viscoelastic medium seismic P-wave and S-wave velocity models, or complex geological structure seismic P-wave and S-wave velocity models. Based on the layout of the VSP data acquisition sensor array (2) and the seismic source (1), it forward models the full-wave field artificially synthesized VSP seismic records recorded in the well by any underground geological structure, as well as the actual VSP data recorded by various existing geophones, and then performs AI training on the VSP data processing artificial big data model (5).
5. The VSP data acquisition system based on an AI artificial intelligence big data model according to claim 1, characterized in that, The VSP data AI processing model (6) is obtained by distillation based on the VSP data processing artificial big data model (5), and is used to process the collected VSP data in the VSP data acquisition and processing computer workstation system (3).
6. A data acquisition and data processing method for the VSP data acquisition system based on an AI artificial intelligence big data model as described in any one of claims 1 to 5, characterized in that, Includes the following steps: (a) Deploy the optimal VSP data acquisition sensor array in the VSP data acquisition work area according to the construction design and downhole conditions (2); (b) Collect three-dimensional geological models, three-dimensional P-wave and S-wave velocity models, acoustic logging data of all wells, VSP data of other wells, rock physical measurement data, anisotropy coefficient and attenuation coefficient Q value in the VSP data acquisition area, and use AI-based underground three-dimensional velocity modeling software to establish three-dimensional viscoelastic medium anisotropic P-wave velocity model and three-dimensional viscoelastic medium anisotropic S-wave velocity model in the VSP data acquisition area (8). (c) Start the VSP data acquisition and processing computer workstation system (3) or the distributed fiber optic acoustic wave sensor DAS modulation and demodulation instrument (4); (d) In the VSP data acquisition work area, start one or more seismic sources (1) to excite at each pre-designed seismic source point, and at the same time collect various VSP data together through the VSP data acquisition and processing computer workstation system (3) or distributed fiber acoustic wave sensor DAS modulation and demodulation instrument (4) and the downhole VSP data acquisition sensor array (2). (e) Simultaneously start the VSP data acquisition and processing computer workstation system (3) which is equipped with the AI real-time VSP data processing model (6) evaporating from the AI-trained VSP data processing artificial big data model (5); (f) The VSP data collected in the VSP data acquisition work area is processed on-site in the VSP data acquisition and processing computer workstation system (3); (g) The VSP data in step (f) is processed by the AI-based VSP data preprocessing software installed in the VSP data acquisition and processing computer workstation system (3) to perform amplitude-preserving noise reduction, surface static correction, surface or downhole consistency processing, deconvolution processing, and AI artificial intelligence wave field separation processing of upper and lower shaped waves. (h) Input the VSP data processed in step (g) and the three-dimensional viscoelastic medium anisotropic longitudinal wave velocity model and the three-dimensional viscoelastic medium anisotropic transverse wave velocity model (8) established in step (b) into the distilled VSP data AI processing model (6); (i) The VSP data AI processing model (6) in the VSP data acquisition and processing computer workstation system (3) obtains the amplitude-preserving high-resolution VSP longitudinal wave imaging data volume and amplitude-preserving high-resolution VSP transverse wave imaging data volume based on the VSP data processed in step (g) and combined with the three-dimensional viscoelastic medium anisotropic longitudinal wave velocity model and the three-dimensional viscoelastic medium anisotropic transverse wave velocity model (8) established in step (b). (j) Extract various VSP longitudinal wave data attributes and various VSP shear wave data attributes from the amplitude-preserving high-resolution VSP longitudinal wave imaging data volume and the amplitude-preserving high-resolution VSP shear wave imaging data volume obtained in step (i). (k) Using the multiple VSP P-wave data attributes and multiple VSP S-wave data attributes extracted in step (j), based on the single attribute data or several combined attribute data that are most sensitive to the comprehensive interpretation target or reservoir physical parameters, perform three-dimensional characterization and fine depiction of the geological structure or oil and gas reservoir spatial distribution within a certain range around the VSP data acquisition well, track the distribution range of oil and gas resources in the oil and gas reservoir around the well, determine the oil / water boundary or gas / water boundary or oil / gas boundary, and calculate the oil and gas saturation in the oil and gas reservoir, thereby achieving a comprehensive evaluation of the oil and gas resource potential around the well and the distribution range of the oil and gas reservoir.