Microseism monitoring system based on AI artificial intelligence big data model and processing method

Abstract

The invention provides a microseism monitoring system and processing method based on an AI artificial intelligence big data model, and the system employs an underground three-dimensional seismic velocity model according to the arrangement mode of microseism signal monitoring sensors. Carrying out forward modeling simulation on full-wave-field artificially synthesized seismic records and micro-seismic data recorded on the ground, the seabed or a well by an underground micro-seismic event, and training a micro-seismic data processing artificial big data model; and evaporating an AI real-time micro-seismic data processing model from the AI-trained micro-seismic data processing artificial big data model. Microseism data collected by a microseism monitoring system is preprocessed, longitudinal wave travel time and transverse wave travel time data of a microseism event are recognized and extracted, an AI real-time microseism data processing model is used for processing, and occurrence time, three-dimensional space position, energy size and fracture mechanism characteristics of the microseism event are obtained. An underground discrete fracture network model is constructed, and long-term dynamic monitoring of the oil and gas well development and production process and the well fluid yield change is achieved.

Description

Technical Field

[0001] The present invention belongs to the application of the AI large model in the field of geophysical exploration technology, and particularly relates to a microseismic monitoring system and processing method based on the AI big data model. Background Art

[0002] Artificial Intelligence (AI) is an important driving force for the new round of scientific and technological revolution and industrial transformation. It is a new technical science that studies, develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. Artificial Intelligence (AI) is an interdisciplinary and emerging discipline that is based on computer science and integrates multiple disciplines such as computer science, psychology, and philosophy. It is a new technical science that studies, develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence, attempting to understand the essence of intelligence and produce a new intelligent machine that can respond in a way similar to human intelligence. The research in this field includes robots, speech recognition, image recognition, natural language processing, and expert systems, etc. Artificial Intelligence is an important part of the intelligent discipline, attempting to understand the essence of intelligence and produce a new intelligent machine that can respond in a way similar to human intelligence. Artificial Intelligence is a very broad science, including robots, speech recognition, image recognition, natural language processing, expert systems, machine learning, computer vision, etc.

[0003] A large model refers to a machine learning model with large-scale parameters and a complex computing structure. These models are usually constructed by deep neural networks and have billions or even hundreds of billions of parameters. The design purpose of the large model is to improve the model's expression ability and prediction performance, and it can handle more complex tasks and data. Large models have a wide range of applications in various fields, including natural language processing, computer vision, speech recognition, and recommendation systems, etc. The large model learns complex patterns and features by training massive amounts of data, and has a stronger generalization ability, and can make accurate predictions for unseen data.

[0004] Big data analysis refers to the process of processing, analyzing and mining large, high-speed, multi-source, and multi-type data to discover valuable information and knowledge. The core technologies of big data analysis include data storage, data processing, data mining, data analysis, and data visualization, etc. With the development of technologies such as the Internet, artificial intelligence, and the Internet of Things, the scale and complexity of data have been increasing continuously, and big data analysis technology has become an indispensable part of enterprises and organizations.

[0005] In the field of deep learning, model compression and deployment is an important research topic, and model distillation is one of the effective methods. Model Distillation was initially proposed by Hinton et al. in 2015. Its core idea is to transfer the knowledge of a complex large model (teacher model) to a relatively simple small model (student model) through knowledge transfer. Briefly, it uses the predicted probability distribution of the teacher model as soft labels to train the student model, thereby greatly reducing the model complexity and computational resource requirements while maintaining high prediction performance, achieving the lightweight and high efficiency of the model.

[0006] Microseismic monitoring technology is a geophysical technology based on acoustics and seismology that monitors the impacts, effects, and reservoir conditions of production activities by observing and analyzing microseismic events generated during production activities. Different from traditional seismic exploration, the location, intensity, and occurrence time of the seismic source are unknown in microseismic monitoring. Determining these unknown factors is the primary task of microseismic monitoring. As a technology developed based on geophysics that can effectively monitor the occurrence location of rock microfractures, microseismic monitoring technology has been widely applied in the fields of mine dynamic disaster monitoring, reservoir stimulation by hydraulic fracturing, etc.

[0007] Fracture microseismic monitoring technology monitors the microseismic waves induced during the fracturing (water injection) process of a fracturing (water injection) well through a downhole three-component geophone array placed in an adjacent well, or a surface single-component or three-component geophone array, or three-component geophones buried in shallow surface wells, to describe the geometric shape and spatial distribution of fractures during the fracturing (water injection) process. It can provide the height, length, and azimuth of the fractures generated during the fracturing operation in real time. Using this information, the fracturing design, well pattern, or other oilfield development measures can be optimized, thereby improving the recovery rate. It is mainly applied in two aspects: fracturing effect evaluation and prediction.

[0008] Microseismic water drive front: Monitor the range and edge of water drive during the water injection process of an injection well by placing three-component geophones in adjacent wells or conducting cross-well seismic surveys. Understand and master the swept area, advancing direction of the injected water of each injection well, and the water swept area of the block, providing a reliable technical basis for the rational deployment of injection-production well patterns, tapping remaining oil, and improving the final economic recovery rate.

[0009] With the rapid development of exploration and development technologies for unconventional resources and the large-scale and widespread application of horizontal well drilling technology and reservoir stimulation technology using hydraulic fracturing, oil and gas companies can now complete drilling, completion, cementing, and hydraulic fracturing operations for up to ten horizontal wells in one wellbore and one well hole at a single well pad at one time. Since there are no other boreholes within several kilometers around the well pad where hydraulic fracturing operations are carried out that can be used for real-time monitoring of microseismic events in adjacent wells during hydraulic fracturing, and because the wellbore where the fracturing operation is in progress cannot be equipped with downhole three-component geophones for real-time monitoring of microseismic events during fracturing in the same well due to the presence of fracturing operation strings, many fracturing operations can only rely on surface single-component or three-component geophone arrays or three-component geophones buried in shallow surface wells to monitor the microseismic events induced during the fracturing process of the fractured well. Summary of the Invention

[0010] The present invention proposes a microseismic monitoring system based on an AI artificial intelligence big data model, including microseismic signal monitoring sensors deployed on the ground, in shallow wells, downhole, or underwater, a ground or shipborne microseismic data acquisition system, a distributed fiber optic acoustic sensing DAS modulation and demodulation instrument, a ground or shipborne microseismic data processing and interpretation workstation, an AI-trained artificial big data model for microseismic data processing, and an AI real-time microseismic data processing model derived from the AI-trained artificial big data model for microseismic data processing.

[0011] The microseismic signal monitoring sensor (1) is connected to the microseismic data acquisition system (2) or the distributed fiber optic acoustic sensing DAS modulation and demodulation instrument (3); the microseismic data processing and interpretation workstation (4) is connected to the microseismic data acquisition system (2) or the distributed fiber optic acoustic sensing DAS modulation and demodulation instrument (3) through a cable; the microseismic data processing and interpretation workstation (4) is provided with an AI-trained artificial big data model (5) for microseismic data processing and an AI real-time microseismic data processing model (6) derived from the artificial big data model (5) for microseismic data processing.

[0012] The microseismic signal monitoring sensor can be one of a velocity geophone, a piezoelectric geophone, an acceleration geophone, an optical fiber geophone, a land wireless node seismograph, a hydrophone, an OBC, an OBN, or an armored optical cable. The armored optical cable is a straight or helical armored optical cable.

[0013] The velocity geophone, piezoelectric geophone, acceleration geophone, and optical fiber geophone can be single-component geophones or three-component geophones; the conventional borehole geophone or optical fiber geophone in the well is a three-component geophone.

[0014] The velocity geophones, piezoelectric geophones, accelerometer geophones, fiber optic geophones, and land wireless node seismographs of the microseismic signal monitoring sensors are arranged on land in a grid pattern or in a radial pattern extending in all directions from the hydraulic fracturing wellhead. Hydrophones, OBCs, or OBNs are arranged on the seabed above the horizontal well section under the sea in a grid pattern or in a radial pattern extending in all directions from the wellhead. The armored straight optical cable or armored spiral optical cable is laid inside and outside the casing or inside and outside the pipe string in the well, and buried in a shallow trench on the land surface or seabed.

[0015] The velocity geophone, piezoelectric geophone, accelerometer geophone, fiber optic geophone, or hydrophone is connected to the microseismic data acquisition system through a cable. The fiber optic geophone or armored optical cable is connected to the distributed fiber optic acoustic sensing DAS modulation and demodulation instrument through the armored optical cable.

[0016] The land wireless node seismograph of the microseismic signal monitoring sensor transmits the recorded microseismic data to the ground microseismic data acquisition system in real time through mobile communication technology, Starlink technology, or low-orbit communication satellite technology; the hydrophones or OBCs deployed on the seabed transmit the recorded microseismic data to the shipborne microseismic data acquisition system through submarine cables in real time, and the OBNs deployed on the seabed transmit the recorded microseismic data to the shipborne microseismic data acquisition system through underwater acoustic communication technology in real time. The microseismic data processing and interpretation workstation is connected to the microseismic data acquisition system through a cable.

[0017] Using underground three-dimensional velocity modeling software based on AI artificial intelligence, inputting 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, rock physics measurement data, anisotropy coefficient, and attenuation coefficient Q value in the microseismic monitoring work area, to establish a three-dimensional viscoelastic medium anisotropic P-wave velocity model and anisotropic S-wave velocity model under the microseismic monitoring work area.

[0018] The AI-trained artificial big data model for microseismic data processing uses various underground three-dimensional seismic velocity models, including homogeneous and inhomogeneous isotropic P-wave and S-wave velocity models, homogeneous and inhomogeneous anisotropic P-wave and S-wave velocity models, acoustic P-wave and S-wave velocity models, elastic wave P-wave and S-wave velocity models, viscoelastic medium P-wave and S-wave velocity models, and complex geological structure P-wave and S-wave velocity models. According to the layout method of the microseismic signal monitoring sensors in the microseismic monitoring work area, forward simulate the full-wavefield synthetic seismic records recorded on the ground, seabed, or in the well for any microseismic event signal underground, as well as the actual microseismic data recorded by various geophones in the past decade or more, to conduct AI training for the artificial big data model of microseismic data processing.

[0019] The described AI real-time microseismic data processing model is obtained by distillation on the basis of the aforementioned artificial big data model for microseismic data processing trained by AI, and is used to perform real-time processing on the collected microseismic data with a microseismic data processing and interpretation workstation on site;

[0020] A processing method for a microseismic monitoring system based on an AI artificial intelligence big data model includes the following steps:

[0021] (1) Deploy an optimal microseismic signal monitoring system in the microseismic signal monitoring work area according to the construction design and underground geological and engineering conditions;

[0022] (2) Collect 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, rock physics measurement data, anisotropy coefficient and attenuation coefficient Q value in the microseismic monitoring work area, and use the underground three-dimensional velocity modeling software based on AI artificial intelligence to establish a three-dimensional viscoelastic medium anisotropic P-wave velocity model and an anisotropic S-wave velocity model under the microseismic monitoring work area;

[0023] (3) Start the microseismic data acquisition system or the distributed fiber optic acoustic sensing DAS modulation and demodulation instrument;

[0024] (4) Implement the acquisition of underground microseismic signals in the microseismic monitoring work area through the microseismic signal monitoring sensor array deployed on the ground or in shallow wells or underground or underwater and the microseismic data acquisition system or the distributed fiber optic acoustic sensing DAS modulation and demodulation instrument;

[0025] (5) Synchronously start the microseismic data processing and interpretation workstation installed with the AI real-time microseismic data processing model evaporated from the microseismic data processing artificial big data model trained by AI;

[0026] (6) Transmit the underground microseismic signals collected in the microseismic monitoring work area to the microseismic data processing and interpretation workstation in real time for real-time processing;

[0027] (7) Use the AI-based microseismic data preprocessing software installed in the microseismic data processing and interpretation workstation to perform amplitude-preserving denoising processing, surface static correction processing, surface or underground consistency processing on the underground microseismic data input in step (6); identify and extract the P-wave travel time data and S-wave travel time data of all underground microseismic events. Calculate the travel time difference between the P-wave and S-wave of the microseismic event; for the S-wave travel time data of the microseismic event for which the P-wave travel time data cannot be extracted, use the P-wave velocity model and S-wave velocity model provided in step (2) to forward calculate the missing P-wave travel time data;

[0028] (8) Input the P-wave travel time data and S-wave travel time data of all microseismic events extracted in step (7) and the three-dimensional viscoelastic anisotropic P-wave velocity model and S-wave velocity model under the microseismic monitoring work area established in step (2) into the distilled AI real-time microseismic data processing model;

[0029] (9) The AI real-time microseismic data processing model in the microseismic data processing and interpretation workstation, based on the P-wave travel time data and S-wave travel time data of the microseismic events obtained in step (7) and the travel time difference between the P-wave and S-wave of the microseismic events, combined with the three-dimensional viscoelastic anisotropic P-wave velocity model and anisotropic S-wave velocity model under the microseismic monitoring work area established in step (2), inversely calculates the occurrence time, three-dimensional spatial position, and energy magnitude of each microseismic event generated during underground formation fracture. Use the P-wave signal characteristics and S-wave signal characteristics of each microseismic event for three-dimensional momentum inversion to obtain the fracture mechanism characteristics of each microseismic event;

[0030] (10) According to the occurrence time, three-dimensional spatial position, energy magnitude, and fracture mechanism of each microseismic event generated during the real-time monitoring of underground formation fracture, observe and analyze the dynamic distribution and changes of the three-dimensional spatial positions of all occurred microseismic events, and analyze the distribution characteristics, laws, and connectivity of tensile fractures, shear fractures, and composite fractures that induce underground microseismic events;

[0031] (11) Calculate the total stimulated reservoir volume SRV of the underground reservoir generated by the hydraulic pressure operation using the envelope of the three-dimensional spatial distribution range of all microseismic events monitored in real-time; According to the distribution characteristics, laws, and connectivity of tensile fractures, shear fractures, and composite fractures and the three-dimensional spatial distribution range of all microseismic events, perform fracture seismic imaging processing based on the source mechanism to generate a discrete fracture network DFN model or an effective stimulated reservoir volume ESRV model induced by underground microseismic events.

[0032] Finally, comprehensively consider the distribution characteristics, laws, and connectivity of tensile fractures, shear fractures, and composite fractures, the total stimulated volume, and the discrete fracture network model or the effective reservoir stimulation model obtained above, and conduct an effective and reliable evaluation of the stimulation effect of unconventional oil and gas resource reservoirs, so as to realize the long-term dynamic monitoring of the oil and gas well development and production process and the change of well fluid production, and provide indispensable means, systems, and methods for the scientific management of oil and gas field development and the improvement of recovery efficiency. Description of the Drawings

[0033] Figure 1 It is a schematic diagram of the process of the microseismic monitoring system and processing method based on the AI artificial intelligence big data model of the present invention.

[0034] Figure 2It is a schematic diagram of the microseismic monitoring system arranged in a grid pattern on the ground or the seabed according to the present invention.

[0035] Figure 3 It is a schematic diagram of the microseismic monitoring system arranged in a radial pattern on the ground or the seabed according to the present invention.

[0036] Figure 4 It is a schematic diagram of the microseismic monitoring system of grid armored optical cables arranged on the ground or the seabed and straight armored optical cables inside and outside the downhole casing according to the present invention.

[0037] Figure 5 It is a schematic diagram of the microseismic monitoring and acquisition system of grid armored optical cables arranged on the ground or the seabed and spiral armored optical cables outside the downhole casing according to the present invention.

[0038] Figure 6 It is a schematic diagram of the microseismic monitoring system of wireless node seismographs arranged on the ground or the seabed according to the present invention.

[0039] Reference numerals in the drawings and corresponding component names: 1 - microseismic signal monitoring sensor, 2 - microseismic data acquisition system, 3 - distributed fiber acoustic sensing DAS modulation and demodulation instrument, 4 - microseismic data processing and interpretation workstation, 5 - AI-trained artificial big data model for microseismic data processing, 6 - AI real-time microseismic data processing model, 7 - three-dimensional viscoelastic anisotropic longitudinal wave velocity model and anisotropic shear wave velocity model, 8 - armored optical cable. Detailed implementation manners

[0040] For the convenience of understanding the present invention, the present invention will be described in more detail below in conjunction with the drawings and specific embodiments. Preferred embodiments of the present 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. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosed content of the present invention more thorough and comprehensive. They do not constitute a limitation on the present invention, but are only examples, and at the same time, the advantages of the present invention will become clearer and easier to understand.

[0041] A specific implementation manner of a microseismic monitoring system and a processing method based on an AI artificial intelligence big data model of the present invention is as Figure 1 shown in the schematic diagram of the process of the microseismic monitoring system and the processing method based on the AI artificial intelligence big data model: including microseismic signal monitoring sensors 1 arranged on the ground, in shallow wells, downholes or underwater, a microseismic data acquisition system 2 on the ground or on a ship, a distributed fiber acoustic sensing DAS modulation and demodulation instrument 3, a microseismic data processing and interpretation workstation 4 on the ground or on a ship, an AI-trained artificial big data model 5 for microseismic data processing, and an AI real-time microseismic data processing model 6 evaporated from the AI-trained artificial big data model 5 for microseismic data processing;

[0042] The microseismic signal monitoring sensor 1 can be one of a velocity geophone, a piezoelectric geophone, an acceleration geophone, an optical fiber geophone, a land wireless node seismograph, a hydrophone, an OBC, an OBN, and an armored optical cable 8. The armored optical cable 8 is a straight or spiral armored optical cable.

[0043] The velocity geophone, piezoelectric geophone, acceleration geophone, optical fiber geophone, and land wireless node seismograph can be single-component geophones or three-component geophones; the downhole conventional geophone or optical fiber geophone is a three-component geophone.

[0044] Such as Figure 2 the schematic diagram of the microseismic monitoring system arranged in a grid pattern on the ground or seabed and Figure 3 the schematic diagram of the microseismic monitoring system arranged in a radial pattern on the ground or seabed as shown, the velocity geophone, piezoelectric geophone, acceleration geophone, optical fiber geophone, and land wireless node seismograph of the microseismic signal monitoring sensor 1 are arranged in a grid pattern ( Figure 2 ) or in a radial pattern in all directions from the hydraulic fracturing wellhead ( Figure 3 ) on land, and hydrophones or OBCs or OBNs are arranged in a grid pattern ( Figure 2 ) or in a radial pattern in all directions from the wellhead ( Figure 3 ) on the seabed above the horizontal well section underground. The armored optical cable 8 is arranged inside and outside the casing or inside and outside the pipe string in the well, and is buried in a shallow trench on the land surface or seabed ( Figure 4 ).

[0045] The velocity geophone, piezoelectric geophone, acceleration geophone, optical fiber geophone, or hydrophone is connected to the microseismic data acquisition system 2 through a cable. Such as Figure 4 the schematic diagram of the microseismic monitoring system with grid armored optical cables arranged on the ground or seabed and straight armored optical cables inside and outside the downhole casing and Figure 5 the schematic diagram of the microseismic monitoring and acquisition system with grid armored optical cables arranged on the ground or seabed and spiral armored optical cables outside the downhole casing as shown, the optical fiber geophone or the armored optical cable 8 is connected to the distributed fiber acoustic sensing DAS modulation and demodulation instrument 3 through an armored optical cable.

[0046] Figure 6It is a schematic diagram of the microseismic monitoring system of wireless node seismographs deployed on the ground or seabed of the present invention. The land wireless node seismograph of the microseismic signal monitoring sensor 1 transmits the recorded microseismic data to the ground microseismic data acquisition system 2 in real time through mobile communication technology, Starlink technology, or low-orbit communication satellite technology; the hydrophone or OBC deployed on the seabed transmits the recorded microseismic data to the shipborne microseismic data acquisition system 2 through a submarine cable, and the OBN deployed on the seabed transmits the recorded microseismic data to the shipborne microseismic data acquisition system 2 through underwater acoustic communication technology. The microseismic data processing and interpretation workstation 4 is connected to the microseismic data acquisition system 2 through a cable.

[0047] Using the underground three-dimensional velocity modeling software based on AI artificial intelligence, input the three-dimensional seismic P-wave velocity model, three-dimensional seismic S-wave velocity model, all acoustic logging data, VSP data, rock physics measurement data, anisotropy coefficient, and attenuation coefficient Q value in the microseismic monitoring work area to establish the three-dimensional viscoelastic medium anisotropic P-wave velocity model and anisotropic S-wave velocity model 7 under the microseismic monitoring work area.

[0048] The AI-trained microseismic data processing artificial big data model 5 uses various underground three-dimensional seismic P-wave velocity models and S-wave velocity models, including homogeneous and inhomogeneous isotropic P-wave and S-wave velocity models, homogeneous and inhomogeneous anisotropic P-wave and S-wave velocity models, acoustic P-wave and S-wave velocity models, elastic wave P-wave and S-wave velocity models, viscoelastic medium P-wave and S-wave velocity models, and complex geological structure P-wave and S-wave velocity models. According to the layout method of the microseismic signal monitoring sensor 1 in the microseismic monitoring work area, forward simulate the full-wavefield synthetic seismic records recorded on the ground, seabed, or in the well of the signals of any microseismic event underground, as well as the actual microseismic data recorded by various geophones in the past ten years, and conduct AI training on the microseismic data processing artificial big data model 5.

[0049] The AI real-time microseismic data processing model 6 is obtained by distillation on the basis of the microseismic data processing artificial big data model 5 trained by AI as described above, and is used to perform real-time processing of the collected microseismic data on-site with the microseismic data processing and interpretation workstation 4;

[0050] The processing method of the microseismic monitoring system based on the AI artificial intelligence big data model includes the following steps:

[0051] (a) Deploy the optimal microseismic signal monitoring system according to the construction design and underground geological and engineering conditions in the microseismic signal monitoring work area;

[0052] (b) Collect the 3D geological model, 3D seismic P-wave velocity model, 3D seismic S-wave velocity model, all acoustic logging data, VSP data, rock physics measurement data, anisotropy coefficient and attenuation coefficient Q value in the microseismic monitoring area. Use the underground 3D P-wave and S-wave velocity modeling software based on AI artificial intelligence to establish the 3D viscoelastic medium anisotropic P-wave velocity model and anisotropic S-wave velocity model 7 underground in the microseismic monitoring area;

[0053] (c) Start the microseismic data acquisition system 2 or the distributed fiber optic acoustic sensing DAS modulation and demodulation instrument 3;

[0054] (d) Implement the acquisition of underground microseismic signals in the microseismic monitoring area through the microseismic signal monitoring sensor 1 array deployed on the ground, in shallow wells, downholes or underwater and the microseismic data acquisition system 2 or the distributed fiber optic acoustic sensing DAS modulation and demodulation instrument 3;

[0055] (e) Synchronously start the microseismic data processing and interpretation workstation 4 installed with the AI real-time microseismic data processing model 6 evaporated from the AI-trained microseismic data processing artificial big data model 5;

[0056] (f) Transmit the underground microseismic signals collected in the microseismic monitoring area to the microseismic data processing and interpretation workstation 4 in real time for real-time processing;

[0057] (g) Use the AI-based microseismic data preprocessing software installed in the microseismic data processing and interpretation workstation 4 to perform amplitude-preserving denoising processing, surface static correction processing, surface or downhole consistency processing on the underground microseismic data input in step (f); Identify and extract the P-wave travel time data and S-wave travel time data of all underground microseismic events. Calculate the travel time difference between the P-wave and S-wave of the microseismic event; For the S-wave travel time data of the microseismic event for which the P-wave travel time data cannot be extracted, use the P-wave velocity model and S-wave velocity model provided in step (b) to forward calculate the missing P-wave travel time data;

[0058] (h) Input the P-wave travel time data and S-wave travel time data of all underground microseismic events extracted in step (g) and the 3D viscoelastic medium anisotropic P-wave velocity model and anisotropic S-wave velocity model 7 underground in the microseismic monitoring area established in step (b) into the distilled AI real-time microseismic data processing model 6;

[0059] (i) The AI real-time microseismic data processing model 6 in the microseismic data processing and interpretation workstation 4 inversely calculates the occurrence time, three-dimensional spatial position, and energy magnitude of each microseismic event generated during underground formation fracturing based on the P-wave travel time data and S-wave travel time data of the microseismic events obtained in step (g) and the travel time difference between the P-wave and S-wave of the microseismic events, in combination with the three-dimensional viscoelastic anisotropic P-wave velocity model and anisotropic S-wave velocity model 7 under the microseismic monitoring site established in step (b). Three-dimensional momentum inversion is performed using the P-wave signal characteristics and S-wave signal characteristics of each microseismic event to obtain the fracture mechanism characteristics of each microseismic event;

[0060] (j) Based on the occurrence time, three-dimensional spatial position, energy magnitude, and fracture mechanism of each microseismic event generated during the real-time monitoring of underground formation fracturing, observe and analyze the dynamic distribution and changes in the three-dimensional spatial positions of all the occurred microseismic events, and analyze the distribution characteristics, laws, and connectivity of the tensile fractures, shear fractures, and composite fractures that induce underground microseismic events;

[0061] (k) Calculate the total stimulated reservoir volume SRV of the underground reservoir generated by the hydraulic pressure operation using the envelope of the three-dimensional spatial distribution range of all the microseismic events monitored in real-time; perform fracture seismic imaging processing based on the source mechanism according to the distribution characteristics, laws, and connectivity of the tensile fractures, shear fractures, and composite fractures and the three-dimensional spatial distribution range of all the microseismic events to generate a discrete fracture network DFN model or an effective stimulated reservoir volume ESRV model induced by underground microseismic events. Finally, comprehensively consider the distribution characteristics, laws, and connectivity of the tensile fractures, shear fractures, and composite fractures, the total stimulated volume, and the discrete fracture network model or the effective reservoir stimulation model obtained above to effectively and reliably evaluate the stimulation effect of unconventional oil and gas resource reservoirs, thereby realizing long-term dynamic monitoring of the oil and gas well development and production process and the change in well fluid production, and providing essential means, systems, and methods for the scientific management of oil and gas field development and the improvement of recovery efficiency.

Claims

1. A microseismic monitoring system based on AI big data model, characterized by: The invention comprises a microseismic signal monitoring sensor (1) arranged on the ground, in a shallow well, underground or underwater; the microseismic signal monitoring sensor (1) is connected to a microseismic data acquisition system (2) or a distributed optical fiber acoustic wave sensing DAS modulation and demodulation instrument (3); the microseismic data processing and interpretation workstation (4) is connected to the microseismic data acquisition system (2) or the distributed optical fiber acoustic wave sensing DAS modulation and demodulation instrument (3) via a cable; The microseismic data processing and interpretation workstation (4) is provided with an AI-trained microseismic data processing artificial big data model (5), and an AI real-time microseismic data processing model (6) evaporated from the microseismic data processing artificial big data model (5); using the AI-based underground three-dimensional velocity modeling software, the three-dimensional geological model, three-dimensional seismic velocity model, all acoustic logging data, VSP data, rock physics measurement data, anisotropy coefficient and attenuation coefficient Q value in the microseismic monitoring area are input to establish a three-dimensional viscoelastic medium anisotropic longitudinal wave velocity model and a transverse wave velocity model (7) underground in the microseismic monitoring area.

2. The microseismic monitoring system based on the AI ​​artificial intelligence big data model according to claim 1 is characterized in that: The microseismic signal monitoring sensor (1) is one of a velocity detector, a piezoelectric detector, an acceleration detector, an optical fiber detector, a land wireless node seismograph, a hydrophone, an OBC, an OBN or an armored optical cable (8); the armored optical cable (8) is an armored straight or spiral optical cable; The velocity detector, piezoelectric detector, acceleration detector, optical fiber detector, and land wireless node seismograph are single-component detectors or three-component detectors; The velocity detector, piezoelectric detector, acceleration detector, optical fiber detector or land wireless node seismograph of the microseismic signal monitoring sensor (1) is arranged on land in a grid manner or in a radial manner from the hydraulic fracturing wellhead in all directions, and the hydrophone or OBC or OBN is arranged on the seabed above the horizontal well section under the seabed in a grid manner or in a radial manner from the hydraulic fracturing wellhead in all directions.

3. The microseismic monitoring system based on the AI ​​artificial intelligence big data model according to claim 2 is characterized in that: The armored optical cable (8) is laid underground inside or outside the casing or inside or outside the pipe column, and buried in a shallow trench dug on the land surface or the seabed.

4. The microseismic monitoring system based on the AI ​​artificial intelligence big data model according to claim 2 is characterized in that: The velocity detector, piezoelectric detector, acceleration detector, optical fiber detector or hydrophone is connected to the microseismic data acquisition system (2) via a cable.

5. The microseismic monitoring system based on the AI ​​artificial intelligence big data model according to claim 2 is characterized in that: The optical fiber detector or armored optical cable (8) is connected to a distributed optical fiber acoustic wave sensor DAS modulation and demodulation instrument (3) via an armored optical cable.

6. The microseismic monitoring system based on the AI ​​artificial intelligence big data model according to claim 2 is characterized in that: The land wireless node seismometer of the microseismic signal monitoring sensor (1) transmits the recorded microseismic data to the ground microseismic data acquisition system (2) in real time through mobile communication technology, Starlink technology or low-orbit communication satellite technology; the hydrophone or OBC deployed on the seabed transmits the recorded microseismic data to the ship-borne microseismic data acquisition system (2) in real time through a submarine cable; the OBN deployed on the seabed transmits the recorded microseismic data to the ship-borne microseismic data acquisition system (2) in real time through underwater hydroacoustic communication technology.

7. The microseismic monitoring system based on the AI ​​artificial intelligence big data model according to claim 1 is characterized in that: The microseismic data processing artificial big data model (5) adopts various underground three-dimensional seismic velocity models, including uniform and non-uniform isotropic longitudinal and transverse wave velocity models, uniform and non-uniform anisotropic longitudinal and transverse wave velocity models, acoustic longitudinal and transverse wave velocity models, elastic longitudinal and transverse wave velocity models, viscoelastic medium longitudinal and transverse wave velocity models, and complex geological structure longitudinal and transverse wave velocity models. According to the layout of the microseismic signal monitoring sensor (1), the full-wave field artificial synthetic seismic record recorded on the ground, seabed or in the well by forward simulation of the signal of any underground microseismic event, as well as the actual microseismic data recorded by various detectors in various places, is used to perform AI training of the microseismic data processing artificial big data model (5).

8. The microseismic monitoring system based on the AI ​​artificial intelligence big data model according to claim 1 is characterized in that: The AI ​​real-time microseismic data processing model (6) is obtained by distillation based on the artificial big data model (5) for microseismic data processing trained by AI; the distilled AI real-time microseismic data processing model (6) is used in the microseismic data processing and interpretation workstation (4) to process the collected microseismic data in real time.

9. The processing method of the microseismic monitoring system based on the AI ​​artificial intelligence big data model according to any one of claims 1 to 8 is characterized in that: The following steps are involved: (a) Deploy the optimal microseismic signal monitoring system in the microseismic signal monitoring area according to the construction design and underground geological and engineering conditions; (b) Collect the three-dimensional geological model, three-dimensional seismic P-wave velocity model and S-wave velocity model, acoustic logging data, VSP data, rock physics measurement data, anisotropy coefficient and attenuation coefficient Q value in the microseismic monitoring area, and establish the three-dimensional viscoelastic medium anisotropic P-wave velocity model and anisotropic S-wave velocity model under the microseismic monitoring area (7); (c) starting a microseismic data acquisition system (2) or a distributed optical fiber acoustic sensor (DAS) modulation and demodulation instrument (3); (d) collecting underground microseismic signals in the microseismic monitoring area through a microseismic signal monitoring sensor (1) array and a microseismic data acquisition system (2) or a distributed optical fiber acoustic wave sensor DAS modulation and demodulation instrument (3) arranged on the ground, in a shallow well, in a well or underwater; (e) synchronously starting a microseismic data processing and interpretation workstation (4) installed with an AI real-time microseismic data processing model (6) evaporated from an AI-trained microseismic data processing artificial big data model (5); (f) transmitting the underground microseismic signals collected in the microseismic monitoring area to the microseismic data processing and interpretation workstation (4) for real-time processing; (g) using the microseismic data processing and interpretation workstation (4) to perform amplitude-preserving denoising, surface static correction, and surface or underground consistency processing on the underground microseismic data input in step (f); identifying and extracting the P-wave travel time and S-wave travel time data of all underground microseismic events; calculating the travel time difference between the P-wave and S-wave of the microseismic events; and for the S-wave travel time data of the microseismic events for which the P-wave travel time data cannot be extracted, forward modeling the missing P-wave travel time data using the P-wave velocity model and S-wave velocity model provided in step (b); (h) inputting the longitudinal wave travel time data and the transverse wave travel time data of all underground microseismic events extracted in step (g) and the three-dimensional viscoelastic medium anisotropic longitudinal wave velocity model and the anisotropic transverse wave velocity model (7) of the microseismic monitoring site established in step (b) into the distilled AI real-time microseismic data processing model (6); (i) The AI ​​real-time microseismic data processing model (6) in the microseismic data processing and interpretation workstation (4) performs inversion calculation based on the longitudinal wave travel time data and transverse wave travel time data of the microseismic event obtained in step (g) and the longitudinal wave travel time difference and transverse wave travel time difference of the microseismic event, combined with the three-dimensional viscoelastic medium anisotropic longitudinal wave velocity model and transverse wave velocity model (7) of the microseismic monitoring site established in step (b), to calculate the occurrence time, three-dimensional spatial position and energy size of each microseismic event generated when the underground stratum is ruptured; the longitudinal wave signal characteristics and transverse wave signal characteristics of each microseismic event are used to perform three-dimensional momentum inversion of the longitudinal wave and transverse wave to obtain the rupture mechanism characteristics of each microseismic event; (j) Based on the occurrence time, three-dimensional spatial position, energy size and rupture mechanism of each microseismic event generated by underground strata rupture monitored in real time, observe and analyze the dynamic distribution and changes of the three-dimensional spatial position of all microseismic events that have occurred, and analyze the distribution characteristics, laws and connectivity of tensile fractures, shear fractures and compound fractures that induce underground microseismic events; (k) Calculate the total transformation volume (SRV) of the underground reservoir generated by hydraulic pressure operation using the envelope of the three-dimensional spatial distribution range of all microseismic events monitored in real time; perform fracture seismic imaging processing based on the source mechanism according to the distribution characteristics, regularity and connectivity of tensile fractures, shear fractures and composite fractures and the three-dimensional spatial distribution range of all microseismic events, and generate a discrete fracture network DFN model or a reservoir effective transformation volume ESRV model induced by underground microseismic events; (l) Using the downhole temperature distribution and fluid noise data measured by the downhole armored optical cable (8), the liquid production profile or water injection profile of the underground reservoir transformation section is calculated, and then the downhole discrete fracture network DFN model or the reservoir effective transformation volume ESRV model is used to compare with the liquid production profile or water injection profile, analyze the correlation between the two, check and confirm the correlation between ESRV and the liquid production profile or water injection profile of the reservoir transformation section, and truly evaluate the effect of reservoir transformation.

Images