Intelligent oil field system based on AI big data model and optical fiber sensing and control method

Abstract

The invention discloses an intelligent oil field system based on an AI big data model and optical fiber sensing and a control method. The intelligent oil field system comprises a multi-parameter optical fiber sensing system, an optical fiber sensing data processing and explaining system and an intelligent oil field system based on the AI big data model, wherein the multi-parameter optical fiber sensing system is arranged on the ground and underground in a three-dimensional mode. The multi-parameter optical fiber sensing system comprises an armored distributed multi-parameter sensing optical cable, an internally-arranged straight armored optical cable, an armored spiral optical cable, an armored distributed three-component DAS optical cable and a distributed multi-parameter multi-component composite optical fiber sensing modulation-demodulation instrument. An artificial intelligence big data model, an underground oil and gas resource high-precision oil reservoir fluid model and distribution characteristics of underground unused or residual or remaining oil and gas resources are utilized, a development and exploitation scheme of an underground oil and gas resource reservoir is optimally designed, an optimal adjustment well or infilled well is arranged, a production system is optimized, and the development and exploitation efficiency is improved. Underground unused or residual or remaining oil and gas resources are developed, the oil and gas yield is increased, and the recovery efficiency of the whole oil and gas field is improved.

Description

Technical Field

[0001] The present invention belongs to the application in the technical fields of artificial intelligence large models and fiber optic sensing technology, and particularly relates to an intelligent oilfield system and control method based on an AI big data model and fiber optic sensing. 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] Artificial intelligence large models generally refer to a class of artificial intelligence models with a large number of parameters constructed by artificial neural networks. Large models are usually trained on a large amount of data through self-supervised learning or semi-supervised learning. Initially, large models mainly referred to large language models (LLMs). With the development of technology, concepts such as vision large models, multi-modal large models, and basic science large models have gradually emerged. Large models are a new concept, and there is no absolutely authoritative definition so far. Therefore, there is no strict standard for the minimum parameter scale that large models need to have. Currently, large models generally refer to models with parameter scales reaching tens of billions, hundreds of billions, or even trillions. In addition, people also habitually refer to models with tens of billions of parameters that have been pre-trained on a large amount of data (significantly more than the training data required by traditional pre-trained models) as large models. A significant feature of artificial intelligence large models is their huge number of parameters. The number of parameters refers to the sum of all trainable parameters in the model, which usually determines the capacity and learning ability of the model. As the number of parameters of the large model increases, it can capture more features and more complex patterns, so it has higher expressiveness when dealing with complex data and learning high-dimensional relationships.

[0004] Large models refer to machine learning models with large-scale parameters and complex computational structures. These models are typically constructed by deep neural networks and have billions or even hundreds of billions of parameters. The design purpose of large models is to improve the model's expressive ability and prediction performance, enabling it to 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. Large models learn complex patterns and features by training on massive amounts of data and have stronger generalization ability, allowing them to make accurate predictions on unseen data.

[0005] 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 continuously increasing, and big data analysis technology has become an indispensable part of enterprises and organizations.

[0006] 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 the way of knowledge transfer. Briefly summarized, it is to use the predicted probability distribution of the teacher model as soft labels to train the student model, thereby greatly reducing the complexity and computational resource requirements of the model while maintaining high prediction performance, achieving the lightweight and high efficiency of the model.

[0007] Large Model Distillation, simply put, is to transfer the knowledge of a complex large model (teacher model) to a smaller model (student model). Just like a teacher imparting their profound knowledge to a student, enabling the student to demonstrate abilities similar to those of the teacher as much as possible under limited resources.

[0008] Large Model Distillation draws on the concept of "knowledge transfer" in the education field and transfers the knowledge of the teacher model to the student model through the way of soft labels. The specific process is as follows:

[0009] Training of the teacher model: First, train a large teacher model to reach a high performance level.

[0010] Knowledge transfer: Using the output of the teacher model (such as probability distribution, intermediate layer features, etc.) as soft labels to guide the learning of the student model. Soft labels not only contain information about the correct class but also carry information about the relationships between classes, enabling the student model to learn knowledge more effectively.

[0011] Optimization of the student model: Through these soft labels, the student model can learn the decision logic and feature representations of the teacher model, thereby improving its performance.

[0012] (1) Methods of large model distillation

[0013] Response-based distillation: The student model learns to imitate the output of the teacher model (only the prediction result), and the teacher model is not trained during the distillation process.

[0014] Intermediate layer feature-based distillation: It not only imitates the output of the teacher model but also imitates the feature representations of its intermediate layers. This method can better transfer the internal knowledge of the teacher model.

[0015] Loss function-based distillation: By adjusting the loss function, the student model can better learn the knowledge of the teacher model.

[0016] (2) Advantages of large model distillation

[0017] Reducing computational cost: The computational complexity of small models is much lower than that of large models. In the inference stage, small models can quickly give results, which is particularly important for applications with high real-time requirements, such as intelligent customer service and voice assistants.

[0018] Reducing storage requirements: Small models have fewer parameters and occupy less storage space, making them easy to deploy in resource-constrained environments such as mobile devices and embedded devices.

[0019] Improving training efficiency: Training small models requires less time and resources, and can complete the training process faster to meet the needs of rapid iteration.

[0020] (3) Application scenarios of large model distillation

[0021] AI applications on mobile devices: Such as intelligent photography and voice recognition functions on mobile phones. Through large model distillation, the energy consumption and computational burden of the device can be reduced without affecting the user experience.

[0022] Edge computing: In Internet of Things devices, due to the limited computing power of the devices, using the distilled small model can directly process data on edge devices, reducing the latency and cost of data transmission.

[0023] Online services: For some online services that need to process a large number of user requests, such as search engines and recommendation systems, using small models can improve the response speed of the service and enhance user satisfaction.

[0024] (4) Challenges in large model distillation

[0025] Knowledge loss: During the knowledge transfer process, the small model may not be able to fully learn all the knowledge of the teacher model, resulting in a decline in performance.

[0026] Optimization of distillation algorithms: How to design more effective distillation algorithms to enable small models to better learn the knowledge of teacher models remains a research hotspot.

[0027] Model adaptation: Different tasks and datasets require different distillation strategies. How to select the appropriate strategy to make the distilled model perform best on specific tasks is also a problem that needs to be solved.

[0028] (5) Latest developments in large model distillation

[0029] DDK framework: In 2024, researchers proposed the DDK (Distilling Domain Knowledge for Efficient Large Language Models) framework, which dynamically adjusts the composition of the distillation dataset according to the performance differences between the teacher and student models in different domains, making the distillation process more stable and effective.

[0030] Distilling Step-by-Step: In 2023, researchers proposed a new distillation mechanism, "Distilling Step-by-Step", which allows training smaller task-specific models with less training data and outperforms few-shot prompting large language models in terms of performance.

[0031] Large model distillation technology provides an effective way to solve the application of large models in resource-constrained scenarios. With the continuous development and improvement of technology, it is believed that large model distillation will play an important role in more fields and promote the wide application of artificial intelligence technology.

[0032] Smart oilfield is a brand-new concept and idea, still in the development stage, and there is no authoritative and unified definition and standard yet. Smart oilfield evolved from digital oilfield, which is an evolution from quantitative change to qualitative change. The core of digital oilfield is digitization and intelligentization, emphasizing artificial intelligence; smart oilfield integrates human wisdom on the basis of digital oilfield, emphasizing the combination of artificial intelligence and human wisdom. Because objects only have intelligence while humans have wisdom. Different from digital oilfield which focuses on data collection, smart oilfield pays more attention to data sorting and the exploration of in-depth application, forming a transformation from "data" to "knowledge". Based on this knowledge, it assists and guides oilfield production decisions, thus optimizing traditional technological processes, providing scientific management methods, and achieving a leap from static to dynamic, intelligent to smart, simple to in-depth, and passive to active. The core idea of smart oilfield is to make full use of information technology means to thoroughly perceive, comprehensively interconnect, deeply intelligentize, and effectively integrate all key information of the core systems of oilfield operation, and make intelligent responses to the needs at all levels such as oilfield production, management, and residents' life, providing scientific and efficient management means for oilfield managers and better living quality for mining area residents.

[0033] 1. Smart oilfield is an oilfield of Internet of Things. A smart oilfield must be an oilfield where things are interconnected. Internet of Things is known as the third information industry revolution following computers and the Internet. It refers to a network system that uses sensing devices such as RFID, sensors, two-dimensional codes, GPS, cameras, etc., connects specific objects to information networks and storage and control systems according to agreed communication protocols for information exchange and automatic control to achieve intelligent identification, perception, positioning, tracking, monitoring, and management. The wide application of Internet of Things in oilfield can enable people to comprehensively perceive things and events in the oilfield, thus "making plans and winning thousands of miles away".

[0034] 2. Smart oilfield is an oilfield with information sharing. Comprehensive, systematic, high-quality, and shareable information is the basis of smart oilfield. Only through sharing can information maximize its value. The more individuals participate in the same information processing and application, the faster the social or economic value of the information grows, and the higher the degree of information sharing.

[0035] Based on the full interconnection of various application systems, using technologies such as information fusion, cloud computing, and fuzzy recognition to achieve regional collaboration and data sharing, and through the analysis and processing of information and data, achieve objective, essential, and comprehensive cognition and judgment, so as to realize the visual and measurable intelligent management and control of the oilfield.

[0036] Data fusion and sharing aims to break the unfavorable situation of "information islands" established separately by each unit and professional department by connecting the dots to form a whole while improving the utilization efficiency of data information, and to establish a scientific and effective intelligent oilfield information sharing mechanism.

[0037] 3. The intelligent oilfield is an oilfield oriented to applications and services. The core of the intelligent oilfield is to establish a new oilfield ecosystem covering oilfield production, management, and residents' lives supported by new tools and technologies. Through the transformation of management concepts and methods, the mode of economic development is changed to achieve leapfrog development from traditional oilfields to emerging oilfields. Therefore, the ultimate goal of the intelligent oilfield is to provide a brand-new management means for oilfield exploration and development, oil and gas production, operation management, and mining area services. Through the application of new methods, the ubiquitous, visual, and intelligent levels in all aspects are improved, and ultimately the green environmental protection and sustainable development of the oilfield are promoted.

[0038] In short, the intelligent oilfield consists of four parts: intelligent geology, intelligent engineering, intelligent management, and intelligent people's livelihood. Summary of the Invention

[0039] To address the above technical problems, the present invention first provides an intelligent oilfield system based on an AI big data model and fiber optic sensing, including:

[0040] A distributed or quasi-distributed multi-parameter fiber optic sensing system deployed comprehensively and three-dimensionally on the ground and underground in the oilfield area;

[0041] A processing and interpretation system that processes, analyzes, statistics, models, and establishes tags for the distributed multi-parameter fiber optic sensing data on the ground and in the wells collected in real time by the multi-parameter fiber optic sensing system;

[0042] An intelligent oilfield system based on an artificial intelligence big data model. The intelligent oilfield system uses all geological structure information, geophysical data, logging data, mud logging data, petroleum engineering data, production data of oil and gas production wells, and injection data of injection wells in the oilfield area to train an intelligent oilfield artificial intelligence big data model, and then distills an AI data processing model for the distributed multi-parameter fiber optic sensing data on the ground and in the wells from the artificial intelligence big data model;

[0043] It also includes an artificial seismic source.

[0044] Further, the multi-parameter fiber optic sensing system includes an armored distributed multi-parameter sensing optical cable, which contains a straight armored optical cable, an armored spiral optical cable, an armored distributed three-component DAS optical cable, and a distributed multi-parameter multi-component composite fiber optic sensing composite modulation and demodulation instrument;

[0045] The straight armored optical cable contains a first single-mode fiber with a high reflection coefficient, a first multi-mode fiber with a high reflection coefficient, a microstructure fiber, or a continuous fiber grating;

[0046] The armored helical optical cable contains a cylindrical elastomer, and a second single-mode optical fiber with a high reflectivity coefficient and a second multimode optical fiber with a high reflectivity coefficient are spirally wound outside the cylindrical elastomer. The angle between the spirally wound second single-mode optical fiber with a high reflectivity coefficient and the second multimode optical fiber with a high reflectivity coefficient and the extending direction of the cylindrical elastomer is at degrees;

[0047] The armored distributed three-component DAS optical cable contains an elastomer with a square or triangular cross-section. On each surface of the elastomer, a third single-mode optical fiber with a high reflectivity coefficient distributed in a sine or triangular shape is fixed, and a straight third single-mode optical fiber with a high reflectivity coefficient along the extending direction of the elastomer is embedded in the center of the elastomer.

[0048] Furthermore, the armored distributed multi-parameter sensing optical cable is buried in a shallow trench on the ground in the oilfield area, and is also laid outside or inside the casing or outside or inside the pipe string of all oil and gas production wells and injection wells in the oilfield;

[0049] The armored distributed multi-parameter sensing optical cable is used to monitor and measure in real time the deformation data of the surface, three-component fluid noise data, three-component strain data, temperature data, three-component SDAS microseismic data, two-dimensional or three-dimensional three-component SDAS seismic data, and three-component time-lapse SDAS seismic data;

[0050] The straight armored optical cable is used to monitor and measure in real time the fluid noise data, strain data, temperature data, seismic data in the formation around the wellbore, downhole microseismic data, and time-lapse vertical surface profile data outside the underground casing or outside the pipe string;

[0051] The armored helical optical cable is used to monitor and measure in real time the fluid noise data, three-component strain data, temperature data, three-component seismic data in the formation around the wellbore, downhole three-component microseismic data, and three-component time-lapse vertical surface profile data outside the underground casing or outside the pipe string;

[0052] The armored distributed three-component DAS optical cable is used to monitor and measure in real time the distributed three-component seismic data, distributed three-component microseismic data, and distributed three-component time-lapse vertical surface profile data on the ground or in the formation around the wellbore.

[0053] Furthermore, the artificial seismic source is one of an explosive seismic source, a controllable seismic source, a drop hammer seismic source, a gas explosion seismic source, an electric energy seismic source, an electric spark seismic source, an air gun seismic source, or a plasma seismic source.

[0054] The artificial seismic source is arranged according to the positions required by the construction design. Zero-offset VSP is excited near the wellhead, or non-zero-offset VSP is excited away from the wellhead, or Walkaway VSP or variable-offset VSP is excited by gradually moving from the wellhead to both ends in sequence, or Walkaround VSP is excited by moving around the wellhead in a circular motion at different radial distances in sequence, or 3D VSP or time-lapse VSP is excited by uniformly arranging in a three-dimensional manner around the wellhead in a grid pattern in sequence, or surface 2D or 3D or time-lapse surface SDAS seismic is excited by uniformly arranging in a two-dimensional or three-dimensional manner within the oilfield area in sequence.

[0055] The present invention also provides a control method for the intelligent oilfield system based on the AI big data model and fiber optic sensing, comprising the following steps:

[0056] (a) Dig shallow trenches on the ground according to the design requirements for the construction of the intelligent oilfield to bury armored distributed multi-parameter sensing optical cables, and arrange armored distributed multi-parameter sensing optical cables outside or inside the casing or outside or inside the pipe string of all oil and gas production wells and injection wells in the oilfield;

[0057] (b) Connect the first single-mode optical fiber with a high reflection coefficient or the second single-mode optical fiber with a high reflection coefficient or the third single-mode optical fiber with a high reflection coefficient in the armored distributed multi-parameter sensing optical cables buried in the ground shallow trenches and arranged in the wells in the oilfield area to the DAS signal input end of the composite modulation and demodulation instrument, connect the first multi-mode optical fiber with a high reflection coefficient or the second multi-mode optical fiber with a high reflection coefficient to the DTS signal input end of the composite modulation and demodulation instrument, connect the first single-mode optical fiber with a high reflection coefficient or the second single-mode optical fiber with a high reflection coefficient to the DSS signal input end of the composite modulation and demodulation instrument, and connect the microstructure optical fiber or continuous fiber grating to the DPS signal input end of the composite modulation and demodulation instrument;

[0058] (c) Connect the armored distributed three-component DAS optical cable in the armored distributed multi-parameter sensing optical cable to the three-component DAS signal input end of the composite modulation and demodulation instrument;

[0059] (d) Establish an intelligent oilfield data collection, storage, backup, processing, analysis, statistics, interpretation, real-time display and control center in the center of the intelligent oilfield, and upload the multi-parameter and multi-component sensing data of the ground and downhole fiber optic sensing networks output by all composite modulation and demodulation instruments to the intelligent oilfield data center in real time through the ground transmission optical cable. In addition, upload the data collected by all ground and downhole electronic sensors and various valves in the intelligent oilfield to the intelligent oilfield data center through the optical cable or cable;

[0060] (e) Training the artificial intelligence big data model of the intelligent oilfield using all the geological structure information, geophysical data, and the processed and interpreted results of logging data and mud logging data, petroleum engineering data, production data of oil and gas production wells, and injection data of injection wells in the oilfield area, and then distilling an AI data processing model for distributed multi-parameter optical fiber sensing data on the ground and in wells from the artificial intelligence big data model;

[0061] (f) Using the artificial intelligence big data model trained in step (e) to construct an accurate underground geological structure model and an oil and gas resource reservoir model of the intelligent oilfield;

[0062] (g) Establishing a high-precision reservoir fluid model of the underground oil and gas resources of the intelligent oilfield using the accurate underground geological structure model and oil and gas resource reservoir model obtained in step (f), as well as the petrophysical parameters and pore fluid parameters in the reservoir;

[0063] (h) Incorporating the three-dimensional coordinates of all the oil and gas production wells and fluid injection wells in the underground of the intelligent oilfield, as well as the fluid outflow and injection data, and the liquid production profile and water / suction gas intake profile data into the high-precision reservoir fluid model of the underground oil and gas resources obtained in step (g);

[0064] (i) Using the artificial intelligence big data model of the intelligent oilfield and the high-precision reservoir fluid model of the underground oil and gas resources after fusion in the method of step (h), as well as the distribution characteristics of the unexploited or residual or remaining oil and gas resources underground, to optimize the development and production plan of the underground oil and gas resource reservoir, arrange the best adjustment wells or infill wells, optimize the production regime, maximize the development of the unexploited or residual or remaining oil and gas resources underground, increase the single-well oil and gas production of the existing oil and gas production wells, and improve the recovery rate of the entire oil and gas field;

[0065] (j) Synchronously, continuously, and long-term collect multi-parameter and multi-component sensing data of the optical fiber sensing network on the ground and downhole in the intelligent oilfield, and use the AI data processing model for distilling distributed multi-parameter optical fiber sensing data on the ground and in wells to process the multi-parameter and multi-component sensing data of the optical fiber sensing network on the ground and downhole in real time;

[0066] (k) In the VSP data acquisition work area of the intelligent oilfield, start one or more seismic sources at each pre-designed seismic source point according to the construction design, and simultaneously collect various time-lapse surface seismic and time-lapse VSP data through a time-lapse surface seismic and time-lapse DAS-VSP data acquisition and processing computer workstation system, a complex modulation and demodulation instrument, and an armored helical optical cable and / or an armored distributed three-component DAS optical cable;

[0067] (l) Perform amplitude preservation processing on the time-lapse surface seismic and time-lapse DAS-VSP data collected in step (k), and extract various surface seismic P-wave data attributes and S-wave data attributes, as well as various VSP P-wave data attributes and S-wave data attributes from the obtained amplitude-preserved high-resolution surface seismic P-wave and S-wave imaging data volumes and amplitude-preserved high-resolution VSP P-wave and S-wave imaging data volumes;

[0068] (m) Utilize the various time-lapse surface seismic P-wave and S-wave data attributes and various time-lapse VSP P-wave and S-wave data attributes extracted in step (l), and based on the single attribute data or several combined attribute data that are most sensitive to the comprehensive interpretation target or reservoir physical property parameters, perform three-dimensional characterization and fine delineation of the geological structure or the spatial distribution of oil and gas reservoirs within a certain range below the surface and around the VSP data acquisition well, track the distribution range and changes of oil and gas resources within the oil and gas reservoirs around the well, determine and track the changes of the oil / water boundary or gas / water boundary or oil / gas boundary, calculate and track the changes of the oil and gas saturation within the oil and gas reservoirs, so as to achieve a comprehensive evaluation of the oil and gas resource potential around the well, the distribution range of the oil and gas reservoir, and the dynamic changes related to exploitation;

[0069] (n) Use the change data of the oil / water boundary or gas / water boundary or oil / gas boundary output within the underground reservoir obtained in step (m), and the change data of the oil and gas saturation within the oil and gas reservoirs to update and adjust in real time the high-precision reservoir fluid model of the underground oil and gas resources obtained in step (h);

[0070] (o) Repeat the process from step (j) to step (n) for the high-precision reservoir fluid model of the underground oil and gas resources updated in step (n), further optimize the development and exploitation plan of the underground oil and gas resource reservoir, deploy new adjustment wells or infill wells, further optimize the production system, maximize the development of the underground unutilized or residual or remaining oil and gas resources, further increase the single-well oil and gas production of the existing oil and gas production wells, and improve the recovery rate of the entire oil and gas field;

[0071] (p) Through the above steps and means, develop and improve the construction of the intelligent oilfield infrastructure, and under the intelligent management of data collection, storage, backup, processing, analysis, statistics, interpretation, real-time display, and control center of the intelligent oilfield, gradually realize the AI artificial intelligence management of the intelligent perception and intelligent control system for the three-dimensional layout on the ground and underground of the intelligent oilfield, and truly realize the comprehensive intelligent operation and intelligent management of the entire process of oil and gas field production;

[0072] (q) Make full use of information technology means to thoroughly perceive, comprehensively interconnect, deeply intelligentize, and effectively integrate the key information of each core system of the oilfield operation, and make intelligent responses to the requirements at each level. Description of the Drawings

[0073] Figure 1It is a schematic diagram of the structure and data processing method flow of the intelligent oilfield system based on the AI big data model and fiber optic sensing of the present invention.

[0074] Figure 2 It is a schematic diagram of the structure of the distributed or quasi-distributed multi-parameter fiber optic sensing system on the ground and downhole of the present invention. Detailed implementation manners

[0075] For ease of understanding the present invention, the present invention will be described in more detail below in conjunction with the accompanying drawings and specific embodiments. The preferred embodiments of the present invention are shown in the accompanying 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 to 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.

[0076] As Figure 1 shown is a schematic diagram of the structure and data processing method flow of the intelligent oilfield system based on the AI big data model and fiber optic sensing of the present invention.

[0077] The intelligent oilfield system based on the AI big data model and fiber optic sensing of the present invention includes a distributed or quasi-distributed multi-parameter fiber optic sensing system 1 arranged comprehensively and three-dimensionally on the ground and downhole in the oilfield area, a processing and interpretation system 2 based on the three-dimensional fiber optic sensing data on the ground and downhole in the oilfield area, and an intelligent oilfield system 3 based on the artificial intelligence big data model.

[0078] Figure 2 Shown is a schematic diagram of the structure of the distributed or quasi-distributed multi-parameter fiber optic sensing system on the ground and downhole of the present invention.

[0079] The multi-parameter fiber optic sensing system 1 includes an armored distributed multi-parameter sensing optical cable 11 arranged on the ground in the oilfield area, which contains a straight armored optical cable 12, an armored spiral optical cable 13, an armored distributed three-component DAS optical cable 14, and a composite modulation and demodulation instrument 15 for distributed multi-parameter multi-component composite fiber optic sensing.

[0080] The straight armored optical cable (12) contains a first single-mode optical fiber with a high reflectivity coefficient (121), a first multi-mode optical fiber with a high reflectivity coefficient (122), a microstructure optical fiber (123), or a continuous fiber grating (124);

[0081] The armored helical optical cable 13 contains a cylindrical elastomer 131. Outside the cylindrical elastomer 131, a second high-reflection coefficient single-mode optical fiber 132 and a second high-reflection coefficient multimode optical fiber 133 are helically wound. The angle between the helically wound second high-reflection coefficient single-mode optical fiber 132 and the second high-reflection coefficient multimode optical fiber 133 and the extending direction of the cylindrical elastomer 131 is about 60 degrees.

[0082] The armored distributed three-component DAS optical cable 14 contains an elastomer 141 with a square or triangular cross-section. On each surface of the elastomer 141, a third high-reflection coefficient single-mode optical fiber 142 with a sinusoidal or triangular distribution is fixed. In the center of the elastomer 141, a straight third high-reflection coefficient single-mode optical fiber 142 extending along the extending direction of the elastomer 141 is inlaid.

[0083] The processing and interpretation system 2 includes processing, analyzing, statistically analyzing, modeling, and establishing tags for all real-time collected ground and downhole distributed multi-parameter fiber optic sensing data. The intelligent oilfield system 3 based on the artificial intelligence big data model includes training an intelligent oilfield artificial intelligence big data model 31 using all geological structure information, geophysical (seismic and non-seismic) data, logging data, mud logging data, petroleum engineering data, production data of oil and gas production wells, and injection data of injection wells in the oilfield area, and then distilling an AI data processing model 32 for ground and downhole distributed multi-parameter fiber optic sensing data from the artificial intelligence big data model 31.

[0084] The armored distributed multi-parameter sensing optical cable 11 is buried in a shallow trench on the ground in the oilfield area and is also laid outside or inside the casing or outside or inside the pipe string of all oil and gas production wells and injection wells in the oilfield. The armored distributed multi-parameter sensing optical cable 11 buried in the shallow trench on the ground in the oilfield area is used to real-time monitor and measure surface deformation data, three-component fluid noise data, three-component strain data, temperature data, three-component SDAS microseismic data, two-dimensional or three-dimensional three-component SDAS seismic data, and three-component time-lapse SDAS seismic data.

[0085] The straight armored optical cable 12 is used to real-time monitor and measure fluid noise data, strain data, temperature data, seismic data in the formation around the wellbore, downhole microseismic data, and time-lapse vertical ground profile data outside the underground casing or pipe string. The armored helical optical cable 13 is used to real-time monitor and measure fluid noise data, three-component strain data, temperature data, three-component seismic data in the formation around the wellbore, downhole three-component microseismic data, and three-component time-lapse vertical ground profile data outside the underground casing or pipe string. The armored distributed three-component DAS optical cable 14 is used to real-time monitor and measure distributed three-component seismic data, distributed three-component microseismic data, and distributed three-component time-lapse vertical ground profile data on the ground or in the formation around the wellbore.

[0086] The ground is also provided with an artificial seismic source 4, which is one of an explosive seismic source, a controllable seismic source, a heavy hammer seismic source, a gas explosion seismic source, an electric energy seismic source, an electric spark seismic source, an air gun seismic source or a plasma seismic source.

[0087] For the intelligent oilfield system based on the AI big data model and fiber optic sensing, the artificial seismic source 4 is arranged according to the position requirements of the construction design, and a zero-offset VSP is excited near the wellhead, or a non-zero-offset VSP is excited away from the wellhead, or a Walkaway VSP or a variable-offset VSP is excited by gradually moving from the wellhead to both ends in sequence, or a Walkaround VSP is excited by moving around the wellhead at different radial distances in sequence, or a three-dimensional VSP or a time-lapse (four-dimensional) VSP is excited by arranging uniformly in a three-dimensional manner around the wellhead in a grid pattern, or a two-dimensional or three-dimensional or time-lapse (four-dimensional) surface SDAS seismic is excited by arranging uniformly in a two-dimensional or three-dimensional manner in the oilfield area in sequence.

[0088] A control method for an intelligent oilfield system based on the AI big data model and fiber optic sensing includes the following steps:

[0089] (a) Dig shallow trenches on the ground according to the design requirements of the intelligent oilfield construction to bury the armored distributed multi-parameter sensing optical cable 11, and arrange the armored distributed multi-parameter sensing optical cable 11 outside or inside the casing or outside or inside the pipe string of all oil and gas production wells and injection wells in the oilfield;

[0090] (b) Connect the first single-mode optical fiber 121 with a high reflection coefficient or the second single-mode optical fiber 132 with a high reflection coefficient or the third single-mode optical fiber 142 with a high reflection coefficient in the armored distributed multi-parameter sensing optical cable 11 buried in the shallow trenches on the ground and arranged in the wells in the oilfield area to the DAS signal input end of the composite modulation and demodulation instrument 15, connect the first multi-mode optical fiber 122 with a high reflection coefficient or the second multi-mode optical fiber 133 with a high reflection coefficient to the DTS signal input end of the composite modulation and demodulation instrument 15, connect the first single-mode optical fiber 121 with a high reflection coefficient or the second single-mode optical fiber 132 with a high reflection coefficient to the DSS signal input end of the composite modulation and demodulation instrument 15, and connect the microstructure optical fiber 123 or the continuous fiber grating 124 to the DPS signal input end of the composite modulation and demodulation instrument 15;

[0091] (c) Connect the armored distributed three-component DAS optical cable 14 in the armored distributed multi-parameter sensing optical cable 11 buried in the shallow trenches on the ground and arranged in the wells in the oilfield area to the three-component DAS signal input end of the composite modulation and demodulation instrument 15;

[0092] (d) Establish a smart oilfield data collection, storage, backup, processing, analysis, statistics, interpretation, real-time display, and control center in the center of the smart oilfield. Transmit the multi-parameter and multi-component sensing data of the surface and downhole optical fiber sensing networks output by all composite modulation and demodulation instruments 15 to the smart oilfield data center in real time through the surface transmission optical cable. In addition, transmit the data collected by all surface and downhole electronic sensors and various valves in the smart oilfield to the smart oilfield data center through optical cables or cables;

[0093] (e) Use all the geological structure information, geophysical (seismic and non-seismic) data, and the processing and interpretation results of logging data and mud logging data, petroleum engineering data, production data of oil and gas production wells, injection data of injection wells, etc. in the oilfield area to train the artificial intelligence big data model 31 of the smart oilfield. Then, distill the AI data processing model 32 of the distributed multi-parameter optical fiber sensing data on the surface and in the well from the artificial intelligence big data model 31;

[0094] (f) Use the artificial intelligence big data model 31 of the smart oilfield trained in step (e) to construct an accurate underground geological structure model and an oil and gas resource reservoir model of the smart oilfield;

[0095] (g) Use the accurate underground geological structure model and oil and gas resource reservoir model obtained in step (f), as well as the petrophysical parameters and pore fluid parameters in the reservoir, to establish a high-precision reservoir fluid model of the underground oil and gas resources of the smart oilfield;

[0096] (h) Integrate the three-dimensional coordinates of all oil and gas production wells and fluid injection wells underground in the smart oilfield, the fluid outflow and injection data, and the liquid production profile and water and gas absorption profiles into the high-precision reservoir fluid model of the underground oil and gas resources obtained in step (g);

[0097] (i) Use the artificial intelligence big data model 31 of the smart oilfield and the high-precision reservoir fluid model of the underground oil and gas resources after fusion by the method in step (h), as well as the distribution characteristics of the underground unused or residual or remaining oil and gas resources, to optimize the development and production plan of the underground oil and gas resource reservoir, arrange the best adjustment wells or infill wells, optimize the production system, maximize the development of the underground unused or residual or remaining oil and gas resources, increase the single-well oil and gas production of the existing oil and gas production wells, and improve the recovery rate of the entire oil and gas field;

[0098] (j) Synchronously, continuously, and long-term collect the multi-parameter and multi-component sensing data of the optical fiber sensing network on the surface and underground in the smart oilfield in real time, and use the AI data processing model 32 of the distributed multi-parameter optical fiber sensing data on the surface and in the well to process the multi-parameter and multi-component sensing data of the optical fiber sensing network on the surface and underground in real time;

[0099] (k) In the VSP data acquisition work area of the intelligent oilfield, start single or multiple seismic sources according to the construction design, and conduct excitation at each pre-designed seismic source point. At the same time, through the time-lapse surface seismic and time-lapse DAS-VSP data acquisition and processing computer workstation system, the complex modulation and demodulation instrument 15 and the armored spiral optical cable 13 and / or the armored distributed three-component DAS optical cable 14 are used together to collect various time-lapse surface seismic and time-lapse VSP data;

[0100] (l) Perform amplitude-preserving processing on the time-lapse surface seismic and time-lapse DAS-VSP data collected in step (k), and extract various surface seismic longitudinal wave data attributes and transverse wave data attributes, and various VSP longitudinal wave data attributes and transverse wave data attributes from the obtained amplitude-preserving high-resolution surface seismic longitudinal wave and transverse wave imaging data volumes and amplitude-preserving high-resolution VSP longitudinal wave and transverse wave imaging data volumes;

[0101] (m) Utilize the various time-lapse surface seismic longitudinal wave and transverse wave data attributes and various time-lapse VSP longitudinal wave and transverse wave data attributes extracted in step (l), and based on the single attribute data or several combined attribute data that are most sensitive to the comprehensive interpretation target or reservoir physical property parameters, conduct three-dimensional characterization and fine description of the geological structure or the spatial distribution of the oil and gas reservoir within a certain range below the ground and around the VSP data acquisition well, track the distribution range and changes of the oil and gas resources in the oil and gas reservoir around the well, determine and monitor the changes of the oil / water boundary or gas / water boundary or oil / gas boundary, calculate and monitor the changes of the oil and gas saturation in the oil and gas reservoir, so as to realize the comprehensive evaluation of the oil and gas resource potential around the well, the distribution range of the oil and gas reservoir, and the dynamic changes related to exploitation;

[0102] (n) Use the change data of the oil / water boundary or gas / water boundary or oil / gas boundary output from the underground reservoir obtained in step (m) and the change data of the oil and gas saturation in the oil and gas reservoir to update and adjust in real time the high-precision reservoir fluid model of the underground oil and gas resources obtained in step (h);

[0103] (o) Repeat the process from step (j) to step (n) for the high-precision reservoir fluid model of the underground oil and gas resources updated in step (n), further optimize the development and exploitation plan of the underground oil and gas resource reservoir, arrange new adjustment wells or infill wells, further optimize the production system, maximize the development of the underground unutilized or residual or remaining oil and gas resources, further increase the single-well oil and gas production of the existing oil and gas production wells, and improve the recovery rate of the entire oil and gas field;

[0104] (p) Through the above steps and means, the construction of the intelligent oilfield infrastructure is developed and improved. Under the intelligent management of data collection, storage, backup, processing, analysis, statistics, interpretation, real-time display and control center in the intelligent oilfield, the AI artificial intelligence management of the intelligent perception and intelligent control system for the three-dimensional layout of the surface and underground of the intelligent oilfield is gradually realized, and the comprehensive intelligent operation and intelligent management of the entire process of oil and gas field production are truly realized;

[0105] (q) Make full use of information technology means to thoroughly perceive, comprehensively interconnect, deeply intelligentize and effectively integrate the key information of the core systems of oilfield operation, and make intelligent responses to the needs at all levels such as oilfield production, management and residents' life, provide scientific and efficient management means for oilfield managers, provide better living quality for residents in the mining area, and build a real intelligent oilfield system.

Claims

1. A smart oilfield system based on an AI big data model and fiber optic sensing, characterized in that, Including: A distributed or quasi-distributed multi-parameter optical fiber sensing system (1) arranged in a comprehensive three-dimensional manner on the ground and downhole in the oilfield area; A processing and interpretation system (2) that processes, analyzes, statistics, models, and establishes tags for the distributed multi-parameter optical fiber sensing data on the ground and in the well collected in real time by the multi-parameter optical fiber sensing system (1); An intelligent oilfield system (3) based on an artificial intelligence big data model. The intelligent oilfield system (3) uses all geological structure information, geophysical data, logging data, mud logging data, petroleum engineering data, production data of oil and gas production wells, and injection data of injection wells in the oilfield area to train an intelligent oilfield artificial intelligence big data model (31), and then distills an AI data processing model (32) of the distributed multi-parameter optical fiber sensing data on the ground and in the well from the intelligent oilfield artificial intelligence big data model (31); It also includes an artificial seismic source (4).

2. The intelligent oilfield system based on the AI big data model and fiber optic sensing according to claim 1, wherein The multi-parameter optical fiber sensing system (1) includes an armored distributed multi-parameter sensing optical cable (11), which contains a straight armored optical cable (12), an armored spiral optical cable (13), an armored distributed three-component DAS optical cable (14), and a composite modulation and demodulation instrument (15) for distributed multi-parameter multi-component composite optical fiber sensing; The straight armored optical cable (12) contains a first single-mode optical fiber with a high refractive index (121), a first multi-mode optical fiber with a high refractive index (122), a microstructure optical fiber (123), or a continuous fiber grating (124); The armored spiral optical cable (13) contains a cylindrical elastomer (131). The second single-mode optical fiber with a high refractive index (132) and the second multi-mode optical fiber with a high refractive index (133) are spirally wound outside the cylindrical elastomer (131). The included angle between the spirally wound second single-mode optical fiber with a high refractive index (132) and the second multi-mode optical fiber with a high refractive index (133) and the extension direction of the cylindrical elastomer (131) is 60 degrees; The armored distributed three-component DAS optical cable (14) contains an elastomer (141) with a square or triangular cross-section. On each surface of the elastomer (141), a third single-mode optical fiber with a high refractive index (142) distributed in a sine or triangular shape is fixed. A straight third single-mode optical fiber with a high refractive index (142) along the extension direction of the elastomer (141) is embedded in the center of the elastomer (141).

3. The intelligent oilfield system based on the AI big data model and fiber optic sensing according to claim 2, wherein The armored distributed multi-parameter sensing optical cable (11) is buried in a shallow trench on the ground in the oilfield area, and is also arranged outside or inside the casing or outside or inside the pipe string of all oil and gas production wells and injection wells in the oilfield; The armored distributed multi-parameter sensing optical cable (11) is used to monitor and measure surface deformation data, three-component fluid noise data, three-component strain data, temperature data, three-component SDAS microseismic data, two-dimensional or three-dimensional three-component SDAS seismic data, and three-component time-lapse SDAS seismic data in real time; The straight armored optical cable (12) is used to monitor and measure fluid noise data, strain data, temperature data outside the underground casing or pipe string, seismic data in the formation around the wellbore, downhole microseismic data, and time-lapse vertical ground profile data in real time; The armored helical optical cable (13) is used for real-time monitoring and measurement of fluid noise data, three-component strain data, temperature data, three-component seismic data in the formation around the wellbore, downhole three-component microseismic data, and three-component time-lapse vertical surface profile data outside the underground casing or outside the pipe string; The armored distributed three-component DAS optical cable (14) is used for real-time monitoring and measurement of distributed three-component seismic data, distributed three-component microseismic data, and distributed three-component time-lapse vertical surface profile data on the ground or in the formation around the wellbore.

4. The intelligent oilfield system based on the AI big data model and fiber optic sensing according to claim 1, characterized in that, The artificial seismic source (4) is one of an explosive seismic source, a vibrator seismic source, a drop hammer seismic source, a gas explosion seismic source, an electric energy seismic source, an electric spark seismic source, an air gun seismic source, or a plasma seismic source.

5. The intelligent oilfield system based on the AI big data model and fiber optic sensing according to claim 1, characterized in that, The artificial seismic source (4) is arranged at the position required by the construction design. It excites a zero-offset VSP near the wellhead, or a non-zero-offset VSP away from the wellhead, or gradually moves from the wellhead to both ends to sequentially excite a Walkaway VSP or a variable-offset VSP, or moves around the wellhead in a circular motion at different radial distances to sequentially excite a Walkaround VSP, or is uniformly arranged in a three-dimensional manner around the wellhead in a grid pattern to sequentially excite a three-dimensional VSP or a time-lapse VSP, or is uniformly arranged in a two-dimensional or three-dimensional manner within the oilfield area to sequentially excite a two-dimensional or three-dimensional or time-lapse surface SDAS seismic on the ground.

6. A control method for the intelligent oilfield system based on the AI big data model and optical fiber sensing according to any one of claims 1 to 5, characterized in that, It includes the following steps: (a) Dig a shallow trench on the ground according to the design requirements of the smart oilfield construction to bury the armored distributed multi-parameter sensing optical cable (11), and arrange the armored distributed multi-parameter sensing optical cable (11) outside or inside the casing or outside or inside the pipe string of all oil and gas production wells and injection wells in the oilfield; (b) Connect the first high-reflection coefficient single-mode optical fiber (121), the second high-reflection coefficient single-mode optical fiber (132), or the third high-reflection coefficient single-mode optical fiber (142) in the armored distributed multi-parameter sensing optical cable (11) buried in the shallow trench on the ground and arranged in the well in the oilfield area to the DAS signal input end of the composite modulation and demodulation instrument (15), connect the first high-reflection coefficient multi-mode optical fiber (122) or the second high-reflection coefficient multi-mode optical fiber (133) to the DTS signal input end of the composite modulation and demodulation instrument (15), connect the first high-reflection coefficient single-mode optical fiber (121) or the second high-reflection coefficient single-mode optical fiber (132) to the DSS signal input end of the composite modulation and demodulation instrument (15), and connect the microstructure optical fiber (123) or the continuous fiber grating (124) to the DPS signal input end of the composite modulation and demodulation instrument (15); (c) Connect the armored distributed three-component DAS optical cable (14) in the armored distributed multi-parameter sensing optical cable (11) to the three-component DAS signal input end of the composite modulation and demodulation instrument (15); (d) Establish a smart oilfield data collection, storage, backup, processing, analysis, statistics, interpretation, real-time display and control center in the center of the smart oilfield. Transmit the multi-parameter and multi-component sensing data of the surface and downhole fiber optic sensing networks output by all complex modulation and demodulation instruments (15) to the smart oilfield data center in real time through the surface transmission optical cable. In addition, transmit the data collected by all surface and downhole electronic sensors and various valves in the smart oilfield to the smart oilfield data center through optical cables or cables; (e) Use all the geological structure information, geophysical data, and the processing and interpretation results of logging data and mud logging data, petroleum engineering data, production data of oil and gas production wells, and injection data of injection wells in the oilfield area to train the artificial intelligence big data model (31) of the smart oilfield, and then distill the AI data processing model (32) of the distributed multi-parameter fiber optic sensing data on the surface and in the well from the artificial intelligence big data model (31); (f) Use the artificial intelligence big data model (31) trained in step (e) to construct an accurate underground geological structure model and an oil and gas resource reservoir model of the smart oilfield; (g) Use the accurate underground geological structure model and oil and gas resource reservoir model obtained in step (f), as well as the petrophysical parameters and pore fluid parameters in the reservoir, to establish a high-precision reservoir fluid model of the underground oil and gas resources in the smart oilfield; (h) Integrate the three-dimensional coordinates of all oil and gas production wells and fluid injection wells underground in the smart oilfield, as well as the fluid outflow and injection data, and the liquid production profile and water / suction profile data into the high-precision reservoir fluid model of the underground oil and gas resources obtained in step (g); (i) Use the artificial intelligence big data model (31) of the smart oilfield and the high-precision reservoir fluid model of the underground oil and gas resources after integration in step (h), as well as the distribution characteristics of the undeveloped, residual or remaining oil and gas resources underground, to optimize the development and exploitation plan of the underground oil and gas resource reservoir, arrange the best adjustment wells or infill wells, optimize the production system, maximize the development of the undeveloped, residual or remaining oil and gas resources underground, increase the single-well oil and gas production of the existing oil and gas production wells, and improve the recovery rate of the entire oil and gas field; (j) Synchronously, continuously and long-term collect the multi-parameter and multi-component sensing data of the fiber optic sensing network on the surface and underground in the smart oilfield, and use the AI data processing model (32) of the distributed multi-parameter fiber optic sensing data on the surface and in the well to process the multi-parameter and multi-component sensing data of the fiber optic sensing network on the surface and underground in real time; (k) In the VSP data acquisition work area of the smart oilfield, start one or more seismic sources at each pre-designed seismic source point according to the construction design. At the same time, collect various time-lapse surface seismic and time-lapse VSP data through the time-lapse surface seismic and time-lapse DAS-VSP data acquisition and processing computer workstation system, the complex modulation and demodulation instrument (15), and the armored helical optical cable (13) and / or the armored distributed three-component DAS optical cable (14); (l) Perform amplitude-preserving processing on the time-lapse surface seismic and time-lapse DAS-VSP data collected in step (k), and extract various surface seismic P-wave data attributes and S-wave data attributes, as well as various VSP P-wave data attributes and S-wave data attributes from the obtained amplitude-preserving high-resolution surface seismic P-wave and S-wave imaging data volumes and amplitude-preserving high-resolution VSP P-wave and S-wave imaging data volumes. (m) Utilize the various time-lapse surface seismic P-wave and S-wave data attributes and various time-lapse VSP P-wave and S-wave data attributes extracted in step (l), and based on the single attribute data or several combined attribute data that are most sensitive to the comprehensive interpretation target or reservoir physical property parameters, perform three-dimensional characterization and fine delineation of the geological structure or the spatial distribution of the oil and gas reservoir within a certain range below the surface and around the VSP data acquisition well. Track the distribution range and changes of the oil and gas resources within the oil and gas reservoir around the well, determine and monitor the changes of the oil / water boundary or gas / water boundary or oil / gas boundary, calculate and monitor the changes of the oil and gas saturation within the oil and gas reservoir, so as to achieve a comprehensive evaluation of the potential of the oil and gas resources around the well, the distribution range of the oil and gas reservoir, and the dynamic changes related to exploitation. (n) Use the change data of the oil / water boundary or gas / water boundary or oil / gas boundary output from the underground reservoir obtained in step (m) and the change data of the oil and gas saturation within the oil and gas reservoir to update and adjust in real time the high-precision reservoir fluid model of the underground oil and gas resources obtained in step (h). (o) Repeat the process from step (j) to step (n) for the high-precision reservoir fluid model of the underground oil and gas resources updated in step (n), further optimize the development and production plan of the underground oil and gas reservoir, deploy new adjustment wells or infill wells, further optimize the production regime, maximize the development of the unexploited or residual or remaining underground oil and gas resources, further increase the single-well oil and gas production of the existing oil and gas production wells, and improve the oil and gas recovery rate of the entire oil and gas field. (p) Develop and improve the construction of the intelligent oilfield infrastructure through the above steps and means. Under the intelligent management of the intelligent oilfield data collection, storage, backup, processing, analysis, statistics, interpretation, real-time display, and control center, gradually realize the AI artificial intelligence management of the intelligent perception and intelligent control system for the three-dimensional layout of the intelligent oilfield on the ground and underground, and truly realize the comprehensive intelligent operation and intelligent management of the entire process of oil and gas field production. (q) Make full use of information technology means to thoroughly perceive, comprehensively interconnect, deeply intelligentize, and effectively integrate all key information of the core systems of the oilfield operation, and make intelligent responses to the requirements at all levels.

Images