Archive

31 December 2024, Volume 43 Issue 6 Previous Issue   

---

For Selected:

Download Citations EndNote Ris BibTeX

Toggle Thumbnails

Select

Research and Thinking on Discipline Construction of Artificial Intelligence in Oil and Gas Industry Liu He, Li Xin, Dou Hong’en, Yan Lin, Wang Hongliang,Liu Junbang, Li Xiaobo, Ren Yili, Li Ning 2024, 43(6): 1-12.  DOI: 10.3969/j.issn.1002-302X.2024.06.001 Abstract ( )   PDF (5208KB) ( )   In order to accelerate the construction of the first-level discipline system of artificial intelligence in the oil and gas industry and the transformation of the talent training model in China, and to solve the contradiction between the shortage of artificial intelligence talents in the oil and gas industry and the uncoordinated demands of society and enterprises, based on the“ Action Plan for Artificial Intelligence Innovation at Universities and Colleges” issued by the Ministry of Education in 2018, the Ministry of Education of China, this article analyzes the global competitive situation of artificial intelligence research and talents, and points out that only by building a good artificial intelligence discipline system can we promote comprehensive improvement of the quality of artificial intelligence talent training in Chinese universities and research institutes, and facilitate the continuous development of teaching and research. It deeply studies the current situation of artificial intelligence discipline construction in 15 domestic and foreign universities and research institutions, and points out that domestic and foreign comprehensive universities, engineering colleges, and research institute schools have not constructed artificial intelligence as a first-level discipline, which lags significantly behind the business and social demands of enterprises, affecting the talent education and training of artificial intelligence. It analyzes the challenges existing in the discipline construction of artificial intelligence in the oil and gas industry in terms of theoretical teaching, practical training, and project practice, proposes the direction and system framework of the talent training model and discipline construction of artificial intelligence in the oil and gas industry, builds a compound talent training system combining “AI+oil and gas business” and “oil and gas business+AI”, updates the talent training model, and promotes the innovative development of the artificial intelligence discipline and the wide application of artificial intelligence technology in the oil and gas industry.

Select

Digitalization and Intelligence Technology Help Refining and Chemical Enterprises Develop towards New Industrialization Wang Hua 2024, 43(6): 13-20.  DOI: 10.3969/j.issn.1002-302X.2024.06.002 Abstract ( )   PDF (4624KB) ( )   To meet the new demand of the digital economic times, China initiated development of new industrialization to optimize and upgrade industrial structure and promote high-quality economic development. This article elaborates the background of new industrialization like data economy and digital transformation and the relationship between intelligent manufacture and new industrialization. Information technology led to digital economy and digital transformation, an inevitable option for adaptation to digital economy. Intelligent manufacture holds the key to realization of new industrialization. It is necessary to continually accelerate integration of digitalization and intelligence technology with the latest manufacturing technology. A number of intelligent factory cases and intelligent application scenarios appeared in the refining and chemical industry after nearly a decade of efforts and practice.Take Sinopec Jiujiang Company, Zhongke (Guangdong) Refining and Chemical Co. Ltd. and PetroChina Guangdong Petrochemical Company for instance. The article analyzes construction and application effects of the domestic typical intelligent refining and chemical plants. It also points out that the future development direction of intelligent plants is to enhance integration of digitalization and intelligence technology with the latest manufacturing technology. Construction should be focused on using new technology like AI and industrial Internet to set up a new generation of intelligent plants based on data drive and human-computer interaction.

Select

Key Problems and Thoughts Related to Application of Intelligent Methods for Oil and Gas Exploration Yang Wuyang, Yang Mowei 2024, 43(6): 21-27.  DOI: 10.3969/j.issn.1002-302X.2024.06.003 Abstract ( )   PDF (5256KB) ( )   There is an unprecedented development opportunity for application of artificial intelligence technology in the oil and gas sector. However, application of this technology still calls for a period of time for study, coordination and accumulation. This article analyzes some key problems facing application of AI technology in the oil and gas exploration sector, such as application scenarios,construction of label data sets, network interpretability, the embedded knowledge of experts and physical mechanism constraints, and intelligent framework and platform. Based on the present conditions of technological development, the article also comes up with the suggestions and solutions related to AI application. Understanding and settlement of those key problems are helpful to mining the data values, optimize the R&D plans of an enterprise, identify the research targets, construct the R&D environment, and facilitate actual application of AI technology.

Select

Scenario and Realization of Logging Artificial Intelligence Application Shi Yujiang, Zhou Jun, Li Xiongwei, Zhang Juan, Chen Yixiang, Li Pengfei, Cui Shitao 2024, 43(6): 28-37.  DOI: 10.3969/j.issn.1002-302X.2024.06.004 Abstract ( )   PDF (12605KB) ( )   Logging is an important link in oil an gas exploration and development. It faces a series of challenges, such as increasingly complicated exploration and development area, serious data problems, incompatible application interfaces, poor generalization of models and weak ability for integration of multi-dimension and multi-size data. In the intelligent and digital transition, China National Logging Corporation aimed at construction of “intelligent and digital CNLC” and focused on three main areas--instrument and equipment intelligent R&D life-cycle management, production and service intelligent control full-process management and integration of intelligent interpretation with logging data. The company enhanced in-depth integration of artificial intelligence with logging technology and strove to establish the intelligent and digital ability to support its business both at the present time and in the future,thus providing reliable technological support for oil and gas exploration and development. Currently, AI application has been realized in a number of scenarios with good results achieved. In the future, the efforts for sustainable development of AI technology will be concentrated on instrument and equipment digital twin application, downhole IOT and integration of multi-dimension and multi-size data.

Select

Practice and Prospect for CNOOC Cloud Computing An Peng, Zhou Bin, Yu Yang, Dong Zhiqiang, Guo Hao 2024, 43(6): 38-45.  DOI: 10.3969/j.issn.1002-302X.2024.06.005 Abstract ( )   PDF (4930KB) ( )   In the era of digital economy, cloud computing has become a new-type service model of digital transformation and intelligent development for oil and gas enterprises. Global oil majors build their corporate digital platform, reduce IT costs and improve utilization rate of resources and efficiency of system development by means of using cloud computing. China National Offshore Oil Corporation (CNOOC) started its cloud computing construction in 2017. Since then, it has kept itself in steps with technological development and built its capability y of“ one cloud for multiple cores, one cloud for multiple stacks, diversified services and coverage of all businesses.” Based on the safety assurance measures and full-process operational system, the offshore oil corporation has ensured security and convenience of cloud use for its businesses, bringing about a sustainable development from “resource cloud” to“ application cloud” and then“ in-depth cloud use.” The cloud servers of CNOOC now exceeded 3500, offering 33 items of cloud services and effectively keeping the company's more than 700 sets of business systems in stable operation. It will focus on consolidation of the platform ability, in-depth integration of businesses and enhancement of operational services at the next step, thus laying a solid foundation for the company's digital transformation and high-quality development.

Select

Large Language Model of Oil and Gas Cognition Constructed and Applied in Shengli Oilfield Duan Hongjie, Ma Chengjie, Wang Zhen, Gong Xuchao, Jing Ruilin, Liu He 2024, 43(6): 46-55.  DOI: 10.3969/j.issn.1002-302X.2024.06.006 Abstract ( )   PDF (10863KB) ( )   To promote the transition of corporate digitalization to intelligence, Shengli Oilfield was committed to accelerate R&D and application of large language model technology in the area of exploration and development. In 2023, it successfully launched the “Shengxiaoli” intelligent assistant online to serve thousands of thousands of users safely. This article briefs about the AI and large language model conception as well as the current development conditions of large language model in the oil and gas sector. Focusing on some bottlenecks in the construction process of the large language model, such as integration and processing of a huge magnitude of multi-source heterogeneous data and a high demand for calculation resources and training efficiency, the “Shengxiaoli” large language model adopts the latest adaptation technology in construction of specialized knowledge, configuration of system technology and optimization of training and reasoning speed. It also learns millions of millions of specialized language materials and knowledge in the oil and gas area, creates the oilfield information assets intelligent service platform based on a large model with billions of natural language resources, and brings about application of high-efficiency exchanged data and sharing of specialized knowledge. Based on operation of the tailored knowledge and construction of the specialized large language model assistant, a dozen of large language models for virtual expert application are established to save users a large amount of time for writing, investigation, consultation of the data system, obviously improving work efficiency. This is a meaningful effort for realization ion of the“ one model for one enterprise and one assistant for one person” goals.

Select

Research and Practice of Smart Gas Field Construction Based on Cloud Native Li Sihai, Qiao Yulu, Xia Qinfeng 2024, 43(6): 56-62.  DOI: 10.3969/j.issn.1002-302X.2024.06.007 Abstract ( )   PDF (4252KB) ( )   Cloud computing technology has developed rapidly in recent years. The latest technology and new-type paradigms are gradually displaying their effectiveness. The cloud-native technology, characterized for microservices, have emerged to accelerate digital transition and satisfy the tailored needs. Sinopec has committed efforts to establishing the Sinopec Smart Cloud industrial Internet platform since it completed the general plan and design of smart oil and gas fields in 2013. Sinopec provided development operations (DevOps) line-style delivery services, improved “business, data and technology” middle-tier capabilities, accelerated transformation of informatization application into components, services and platforms and supported oil and gas field construction with platform and technology. Based on the Sinopec Smart Cloud platform, Fuling smart shale gas field of Jianghan Oilfield Company was constructed according to the new model of “data+platform+application,” covering seven main business scenarios--dynamic gas reservoir management, optimization of production operation, single-well management, equipment management, manifold management, HSE and collaborative study. Improvement of cloud native development capability and reduction of the difficulties and cost for development, testing, deployment, and operation of the information system dramatically raised the digitalized information level of production and management, providing good references for digital transition of oilfield enterprises.

Select

Design and Practice of Intelligent Longwangmiao Gas Field Wang Zhouyang, Zhao Chenyang, Lai Wenhua, Zou Zihan, Jiang Dawei 2024, 43(6): 63-71.  DOI: 10.3969/j.issn.1002-302X.2024.06.008 Abstract ( )   PDF (17541KB) ( )   This article elaborates the development process of domestic and foreign intelligent gas fields, clarifies the core framework and key technology of intelligent gas field construction and analyzes the four typical characteristics of intelligent oil and gas fields--full sensing, automatic operation and control, prediction of the trend and optimization of decision-making. It also focuses on the production problems appearing in the middle stage of Longwangmiao gas reservoir development on Moxi Block of Sichuan Basin,such as declining formation ability, rising water invasion and difficulties for keeping production stable. The efforts are made for establishment of the harmonious support environment and use of the gas reservoir-wellbore-surface integrated simulation and intelligence work flow technology to construct the environment confirming the intelligent Longwangmiao gas field and establish the intelligent allocation work flow and intelligent tracing and diagnosis work flow to realize digital characterization of oil and gas field assets. From a panoramic angle, the article describes the production system, the real-time optimization of development and production, re-construction of the information-based high-efficiency oil and gas field development management flow, acceleration of “data-sharing, flow automation, business coordination and decision-making intelligence” in the development and production process, thus bringing Longwangmiao gas reservoir under high-efficiency development and continually stable production.

Select

Practice and Thinking of Constructing Data Treatment Tools-- Taking CNOOC for Instance Gao Jianyi 2024, 43(6): 72-80.  DOI: 10.3969/j.issn.1002-302X.2024.06.009 Abstract ( )   PDF (5390KB) ( )   Data treatment is a difficult and complicated task calling for continuous efforts. Data treatment tools play an important role in effectively improving work efficiency of data treatment, providing professionals with high-quality data services and making data “visible, understandable, usable and operable.” This article focuses on the bottlenecks in construction of data treatment tools, such as complicated demands, variety of metadata and integrated communication. Combined with CNOOC data treatment work and data treatment tool construction practice, the article makes detailed introduction of the construction methods for key functions and the achievements in the area, such as data asset catalogue, data standards management, data model management, metadata management,data quality management, data safety management and data service management. It also proposes to establish a set of systematic data treatment tools to help enterprises so that they can tap the value of data elements in their efforts for continual improvement of data management work.

Select

Intelligent Transformation Research and Practice of Shengli Offshore Oilfield Development Yuan Xiangbing, Dong Yan, Li Shouqin 2024, 43(6): 81-87.  DOI: 10.3969/j.issn.1002-302X.2024.06.010 Abstract ( )   PDF (4222KB) ( )   The production parameters collected by Shengli offshore oilfield is compete on the basis of production information and database. With development unfolded continually, a large quantity of data came under accumulation. It is of great urgency to uncover data values, rapidly find the anomalies, optimize injection and production plans, tap reservoir potentials and improve work efficiency.Shengli offshore oilfield has been committed d to“ comprehensive perception, integration-based synergy, early warning and prediction,and analysis and optimization” since the start of the 13th Five-year Plan, thus accelerating intelligent transformation of oilfield development. Currently, oilfield development management was concentrated on production analysis, adjustment of injection and production, dynamic analysis, and inspection of indexes. The study also led to some other functions, such as automatic reminder of reservoir anomalies, recommendation and optimization of the adjustment plans, and analysis and tracing of the production adjustment plans. As a result, the closed management model for reservoir development was able to“ find problems, formulate measures, follow up results and evaluate performance,” thus accelerating transition of reservoir management to reservoir analysis and making decisions on business.

Select

Research an Development of Domestic Geophysical Software Platform and Construction of Ecological System Sun Xiaoping,Wen Jiamin,Shang Minqiang,Lei Na,Du Jiguo,Zhou Kun 2024, 43(6): 88-95.  DOI: 10.3969/j.issn.1002-302X.2024.06.011 Abstract ( )   PDF (12761KB) ( )   GeoEast is the main domestic geophysical software. To enhance construction of the ecological system is an inevitable development way for improving technological R&D efficiency, accelerating transformation of the achievements and bringing about reciprocation of the industrial advantages. GeoEast-iEco, the all-new generation of software platform, provides a series of multi-level, deregulated and integral services, including high-efficiency exploration data management and software development support. With the progress continually made in the technological area, intelligent development operation support, convenient cloud transformation, domestic hardware and software adaptation, and the windows-supported cross-platform development have laid a solid foundation and a technological basis for creating the independently controllable geophysical software ecology. Based the iEco platform, the coconstruction, sharing and win-win mechanism is tried and established in the efforts to support construction of the ecological system while continually accelerating a series of services, actively fostering external R&D strength, establishing the standards and specifications, and launching the demonstrative projects. The next step is to continue perfection and development of the ecological system in the direction of intelligence, ecologization, independence and cloudification.

Select

Thoughts and Prospects on The Intelligent Transformation of Reservoir Simulation Wu Shuhong, Bi Jianfei, Fan Tianyi, Xu Mingyuan, Yang Yixuan, Wang Baohua, Li Hua, Li Qiaoyun 2024, 43(6): 96-106.  DOI: 10.3969/j.issn.1002-302X.2024.06.012 Abstract ( )   Reservoir simulation, as a key technology for oil and gas reservoir development and management, is crucial for enhancing recovery in mature fields and increasing reserves and production in unconventional resources. In recent years, breakthroughs in deep learning have provided new approaches and insights for improving the accuracy and efficiency of reservoir simulation, driving its intelligent transformation. This article delves into the applicable scenarios and latent value of diverse deep learning models, including pure data-driven models, operator learning models, and physics-informed models, within the domain of reservoir simulation. It also examines the research progress of artificial intelligence in facilitating reservoir numerical simulation and constructing efficient surrogate models, clarifying the application prospects and limitations of deep learning in reservoir simulation. The research proposes that the future development of intelligent reservoir simulation will concentrate on directions such as constructing specialized large models, exploring the solution of the partial differential equation (PDE) for fluid flow, and developing agent-based technologies.

Select

Application of Artificial Intelligence Pretrained Foundation Models in Exploration and Development and Challenges in This Area Yang Minghao,Li Xiaobo,Liu Xingbang,Zeng Qian 2024, 43(6): 107-113.  DOI: 10.3969/j.issn.1002-302X.2024.06.013 Abstract ( )   PDF (4073KB) ( )   Artificial intelligence (AI) has developed towards the powerful stage characterized for pretrained foundation models from the weak one since ChatGPT was released at the end of 2022. With the capability of pretrained foundation models rising rapidly, actual application of industrial scenarios has become a focus of concerns from the production, investment and research communities.Base on investigation of languages, images and multi-model state, pretrained foundation models can be used in the oil and gas exploration and development area for knowledge management, seismic interpretation, experimental image analysis, generation of data analysis and predictive maintenance of equipment. Obviously, the demand for application of AI pretrained foundation models is huge in the oil and gas exploration and development area. Influenced by the characterizations of oil and gas business and data, actual application of AI pretrained foundation models still faced some challenges, such as integration of specialized data, depth of domain understanding and safety for application of pretrained foundation models. With the new progress made in the areas of multi-model state capability, context learning, Agent and embodied artificial intelligence, AI pretrained foundation model technology will hopefully improve the ability for settlement of complicated tasks in the oil and gas exploration and development area and accelerate digital transformation and intelligent development of the oil and gas upstream business.

Select

Knowledge Graphs: Key Technology for Enhancing RAG Performance in the Oil and Gas Industry Song Ziyu 2024, 43(6): 114-125.  DOI: 10.3969/j.issn.1002-302X.2024.06.014 Abstract ( )   PDF (10281KB) ( )   Addressing the limitations of traditional Rule-Augmented Generation (RAG) in aspects such as associative analysis, information integration, and logical reasoning capabilities, this study focuses on the integration methods of knowledge graphs and RAG by analyzing the research progress and application cases at home and abroad. ChatLaw introduces domain experts to precisely define legal entities, relationships, and cases, enhancing the accuracy of legal consultations through high-quality knowledge graphs. GraphRAG employs knowledge graphs to represent entities and relationships in unstructured text, leveraging techniques such as hierarchical clustering and summarization generation to improve RAG’s global search capabilities on large-scale datasets. HippoRAG utilizes knowledge graphs for concept expansion and retrieval during the query step, enhancing RAG’s knowledge integration and multi-hop reasoning abilities. This study summarizes fusion methods of RAG and knowledge graphs, suggesting that incorporating knowledge graphs in stages such as data partitioning, data storage, query optimization, retrieval recall, reranking, prompt construction,and answer generation can improve RAG’s accuracy, associative analysis capabilities, reasoning abilities, and interpretability. Based on open-source frameworks such as Lucene and LangChain, three retrieval schemes—full-text retrieval, vector retrieval, and graph retrieval—are designed and applied to oil and gas knowledge Q&A scenarios, validating the effectiveness of knowledge graphs in enhancing RAG.