The development trend of AI in the oil and gas industry mainly includes the following aspects: First, improving application efficiency. With the development of the oil and gas industry, the demand for productivity is also increasing. AI technology plays an important role in this field. For example, machine learning can effectively improve work efficiency and help enterprises respond to market changes quickly, Flexibility to meet customer needs. 2、 Improving maintenance and management Using AI technology, we can more effectively maintain and manage equipment and facilities in the oil and gas industry, reduce maintenance costs, improve equipment reliability and service life, and improve production efficiency and level in the oil and gas industry. 3、 To realize safety monitoring, using artificial intelligence technology, can realize efficient monitoring of oil and gas industry detection security, effectively solve emergency security incidents, and at the same time make scientific and reasonable safety warnings for the industry to ensure the safe development of the industry. 4、 Improving customer service Using AI technology, we can realize intelligent customer service, intelligent customer service and other functions in the oil and gas industry, improve customer experience, and improve the competitiveness of the oil and gas industry.
Industry observation
AI+upstream, how to break the situation
Artificial intelligence (AI) technology is known as the engine of the "fourth" industrial revolution, which will greatly promote the digital transformation and intelligent development of the oil and gas industry, and produce significant social and economic benefits. Tress data shows that in 2019, more than 3000 oil and gas companies will spend about $1 trillion on the operation of oil wells and related infrastructure. If the automation and digitization process is accelerated, the expenditure can be reduced by about 10%. PwC predicts that by 2025, the upstream business of oil and gas companies will save $100 billion to $1 trillion in capital and operating expenses through the application of artificial intelligence technology.
In recent years, the oil and gas industry has faced the problem of low oil prices. Many international oil companies and oil service companies have joined hands with IT giants to achieve cross-border cooperation, strengthen the construction of artificial intelligence platforms, promote the rapid digital transformation of exploration and development business, and effectively enhance the potential for efficiency. In 2017, Schlumberger cooperated with Google to launch the DELFI cloud platform, deeply integrating big data, cognitive computing and other technologies with oil and gas exploration and development and other businesses, building a digital, automated and intelligent professional application environment for the whole process of exploration and development, supporting the innovative development of enterprise transformation, and making the platform development enter the era from "N" to "1".
Since 2020, major oil companies have stepped up efforts to build smart oil and gas fields to reduce costs and improve efficiency. Shell proposed to build a smart oilfield, with the goal of increasing production by 10%, improving oil recovery by 5%~10%, reducing operating costs by 20% and shortening oilfield development cycle by 50%.
At present, the construction level of China's intelligent oil fields is in the transition stage from digital to intelligent. A few oil regions have basically completed the prototype of intelligent oil fields, which have been equipped with self diagnosis, early warning and alarm of oil wells, and can recommend optimized decision-making schemes.
Challenges to AI development in oil and gas industry
In the past 20 years, China's oil enterprises have created the "Internet of Things construction mode of oil and gas fields" and "intelligent oil and gas field construction mode" in the construction of intelligent oil and gas fields. The integrity, scale and management of intelligent oil and gas fields have been in the forefront of foreign oil and gas fields. However, the development of human intelligence in the oil and gas industry still faces a series of challenges.
First, the challenges brought by such problems as exploration and development data islands. In the past decades of upstream exploration and development, problems such as multiple data entry, inconsistent standards, repeated development of functions, and loose integration of information and business have gradually emerged, resulting in more databases, more platforms, and more isolated applications. Data sharing and business collaboration are difficult, posing challenges to the landing and application of AI in the oil and gas industry. The landing and application of high-quality AI technology requires high-quality big data as the premise and foundation. Due to the limited acquisition technology, the data that can represent the characteristics of the problem do not have diversity characteristics, and the big data with single characteristics is not really big data, which also brings challenges to the research in the complex oil and gas exploration and development field. Data is the soul of AI technology development. Big data, data quality and its governance determine the future of intelligent development. Grasping data and data governance in digital transformation is equal to grasping the future of AI technology development.
Second, the barrier challenge between AI algorithm engineers and business personnel. In general, business personnel do not understand AI algorithms, and algorithm engineers also do not understand the expertise of business personnel. As a result, the phenomenon of "incomprehensible, unclear, and incoherent" often exists between algorithm engineers and business engineers, which brings obstacles to the implementation of AI in the business field. The application of AI in the oil and gas industry is different from other industries. The objects of oil and gas exploration and development are all underground, which is an invisible black box system. AI technology is used to solve problems. Different from the AI AlphaGozero in human-computer game, the chess board it follows is regular and visible. Most of the problems to be solved and dealt with in the oil and gas industry are not subject to any rules or problems, and have strong uncertainty. To solve the AI application problems in the oil and gas industry, professional knowledge and industry experience are very important, and the challenges of barriers between algorithm engineers and business personnel need to be solved. Only the deep integration of both sides is needed, Only in this way can AI technology products or scenarios be applied.
Third, the continuous investment of funds brought about by the continuous iterative development of intelligent application scenarios. The research on the basic theory and technical principle of artificial intelligence technology and intelligent oil and gas field construction is not deep enough, the technology and methods are not mature enough, and there is no fully formed reference model at home and abroad. In the process of artificial intelligence scene and intelligent oil and gas field construction, there are six missing links of data acquisition, data transmission, data storage, data processing, data management and data use to varying degrees. Some oilfields have only acquisition or video equipment, and data and image analysis technology cannot keep up, leading to "more built, less used" to varying degrees Or the phenomenon of "only building but not building".
Therefore, the currently built AI application scenarios and intelligent oil and gas fields are relatively preliminary. The emergence of these phenomena has also affected the process of digital and intelligent transformation to a certain extent. Artificial intelligence technology scenarios and intelligent oil and gas field construction are not like building construction, and completion acceptance can be completed. It needs to continuously invest upgrading funds according to technological progress and innovative development, constantly adapt to changing business needs and user experience, and constantly update iteratively. The research, planning and deployment of AI application scenarios and intelligent oilfield construction should be comprehensive. However, as far as the strategy of AI application scenarios and intelligent oilfield construction is concerned, it is necessary to focus on, break through point by point and line by level, and finally achieve comprehensive intelligence.
Fourthly, AI high-end technologies and products are restricted by foreign countries. Artificial intelligence technology development and application scenarios are implemented. More than 50% of the technologies and products used in the construction of intelligent oil and gas fields, such as high-performance intelligent sensors, cloud servers, cloud computing software, are from Europe and the United States. However, in recent years, it is difficult to introduce high-end technology, which has brought challenges to the artificial intelligence of the oil and gas industry and the construction of intelligent oil and gas fields. In recent years, although China has caught up with the development of high-end AI technology, there is still a gap between the performance of a few products and that of Europe and the United States. State owned oil and gas enterprises need to solve the dilemma of key core technologies in the short term.
Fifthly, challenges brought by talent shortage. Artificial intelligence technology and application and intelligent oil and gas field construction not only need a group of technicians who understand data science, network operation technicians, and senior programmers, but also need compound talents who understand both oil and gas business and artificial intelligence. At present, comprehensive universities need to set up AI related majors, and enterprises need to set up corresponding positions and professional title sequences; The landing of AI scenarios can help universities, high-tech companies, labeling companies, software companies and other social research and development forces to form a joint research team to solve the problem of talent shortage in the transformation of digital AI.
Suggestions on AI development in oil and gas industry
First, set up a key laboratory of artificial intelligence to accelerate the development and incubation of high-end technology products.
Build a key laboratory of artificial intelligence, carry out data intelligence experiments, intelligent computing and intelligent platform research and development of the whole business chain of exploration and development, organically combine cloud computing, big data, Internet of Things, mobile Internet, artificial intelligence and blockchain and other technologies with the main business, realize the transformation and upgrading of traditional oil and gas industry, and jointly build and share, business collaboration, network interconnection, data exchange Intelligent decision-making and ecological reconstruction. Taking the construction of key AI laboratories as a starting point, we will open up the whole business chain of exploration and development engineering data, create a hybrid cloud platform, accelerate the research and development and incubation of high-end AI products, help improve exploration and development efforts, and play a good exploration and development offensive.
Second, the cross integration of AI technology and business improves the technology support to lead the future.
As a general technology, AI technology will touch all fields of the oil and gas industry in the future. In a real sense, it will fully realize the highly cross integration of AI and traditional business, and become a new industry in the oil and gas field - intelligent oil and gas. In the next 10 to 15 years, the upstream key business development of the intelligent oil and gas industry will aim at five major fields, namely, intelligent exploration, intelligent development, intelligent engineering, intelligent production operation and optimization decision-making, and intelligent data governance. Focus on tackling four key technologies. First, carry out the tackling and application of computer vision application technology, knowledge mapping application technology, machine learning based application technology and other basic technologies in the field of oil and gas exploration and development, and innovate and break through the basic key technologies of artificial intelligence; Second, accelerate the research on intelligent exploration technologies such as intelligent processing of seismic data, intelligent prediction of lithofacies, sedimentary facies and geological desserts, and intelligent evaluation of exploration targets, so as to create efficient and accurate intelligent evaluation technologies for oil and gas targets; Third, strengthen the research on technologies such as geological modeling of intelligent oil and gas reservoirs, intelligent simulation of physical and data driven reservoirs, intelligent optimization of geological reservoir engineering integration, intelligent injection and production of gas reservoirs, and innovate to form digital twin oil and gas development technologies; The fourth is to carry out technical research on the standardization of oil and gas exploration and development data, the application of data trusted security management, data governance and sharing, and data marts, so as to create intelligent data in the upstream field of oil and gas.
Third, strengthen the training and introduction of compound talents.
As AI and petroleum exploration and development cover a wide range of disciplines, it is difficult to cultivate complex talents and the period is long, so it is necessary to vigorously cultivate, introduce and employ AI young leading talents and teams in a variety of ways. Formulate talent training plans of relevant research institutes, and incline to the field of artificial intelligence; Joint cultivation of key universities with AI majors; With the help of talent introduction plan, strengthen the introduction and cultivation of high-end AI talents, and introduce and recruit talents with dual professional backgrounds of AI and oil and gas from home and abroad. Strengthen multi-party cooperation, school enterprise cooperation, in-depth cooperation between petroleum enterprises and IT enterprises to cultivate interdisciplinary talents, establish interdisciplinary joint research teams, achieve cross-border integration, carry out digital and intelligent technology research on exploration and development business chain, and truly play the role of "production, learning, research and application".
Fourth, quickly introduce precise support policies to encourage innovation and transformation.
AI is a major opportunity for China to catch up with the scientific and technological frontier and even lead innovation. In order to accelerate the AI innovation led development of China's oil enterprises, more effectively support the innovative development and scenario landing test of the new generation of AI in oil and gas exploration and development, build an AI original center, seize the commanding heights of AI technology in the oil and gas industry, and propose policy support and innovation incentive policies, Establish an AI technology R&D and innovation fund or industrial innovation fund for oil enterprises, and provide strong financial support and policy incentives for original AI projects.
Deepen the reform of scientific research management and scientific and technological investment system and mechanism, establish early investment, long-term investment, phased continuous investment and industrial chain portfolio investment mechanism for AI industry, establish a highly operational and implementable AI scientific research investment incentive mechanism, stimulate the R&D innovation vitality of R&D personnel, and make more achievements for scientific researchers Make quick achievements and create good scientific research conditions.
Set up major digital intelligence projects corresponding to each business field of oil and gas to ensure that all key technologies of digital intelligence can have corresponding R&D investment, so that there is no dead end in the transformation and development of digital intelligence. Through special research, we will comprehensively realize the research and development of new technologies, new products and new processes, form new industries and new formats, and achieve industrial transformation. (Li Xindou Hongen, Artificial Intelligence Research Center of China Petroleum Exploration and Development Research Institute)
Oil Sea Tide Observation
Data+upstream, how to go
Data is a key factor of production in the digital economy era, a key achievement of the third industrial revolution, and an important foundation of the fourth industrial revolution. Data assets have become the core competitiveness of enterprise development. It is imperative to strengthen the precipitation and construction of data assets. With the acceleration of digital transformation process, data science and big data technology have become the core engine for digital transformation and intelligent development of related industries.
Applying data science and big data technology, mining the value of data assets in the oil and gas field, and providing efficient data and integrated services to support scientific research and decision-making management in the oil and gas exploration and development field are of great significance in promoting the digital transformation of the oil and gas field, which is the only way for China's oil and gas enterprises to become first-class international oil companies.
New opportunities for digital transformation of oil and gas enterprises
General Secretary Xi Jinping once stressed: "We should seize the opportunities given by industrial digitalization and digital industrialization to accelerate the construction of new infrastructures such as 5G networks and data centers." On April 20, 2020, the National Development and Reform Commission explicitly included the data center as an information infrastructure in the category of "new infrastructure". With the national emphasis on strategic emerging industries and the proposal of the task of "new infrastructure", the data center has ushered in new development opportunities. In the future, several national data centers will be built one after another. It is estimated that the scale of China's data native industry will account for 15% of the total economic volume in 2030, and the overall scale of data will exceed 4YB, accounting for 30% of the total global data volume. Data science and big data related applications will enter the rapid development track.
China's oil and gas exploration is in the middle stage of exploration as a whole. In recent years, the newly discovered oil and gas reserves and resources of large scale are mainly concentrated in ultra-low permeability, deep and unconventional fields. How to apply data science and big data technology to improve the accuracy of structural interpretation, the coincidence rate of reservoir interpretation, the success rate of geological target drilling, etc., is to consolidate the long-term stable production of the company's crude oil It is an important means for the sound growth of natural gas. At present, China's oil enterprises have built a series of information management systems to effectively manage and apply structured data. However, in view of the lack of management systems for various reports, articles and achievements, data assets have not been really established, and massive data still cannot be open and shared to effectively meet business applications. The integrated application of data science and big data technology with oil and gas exploration and development business will provide new opportunities for digital transformation, high-quality development in the oil and gas field and the realization of enterprise strategic objectives.
Data+upstream application roadmap
At present, the application of data science and big data technology has made some achievements, but also faces many challenges. First, the application of data science and big data technology needs high-quality and all-round data support, and data governance is crucial; Second, the scattered data in the upstream field of oil and gas still need to be deeply integrated, especially the research and use of new generation information technology such as knowledge atlas to achieve the fusion of multi-source heterogeneous data, and then build a complete knowledge system; Third, there is no unified platform management and control for different business applications and business applications and data of different organizations, which makes it difficult to mine the value of data from the global level. You can build services through the data middle office, establish professional data associations, and realize the interoperability, sharing and reuse of applications and data. Therefore, the construction of data governance, knowledge map and data middle platform in the exploration and development field will become the core of the digital transformation in the upstream field.
First, attach importance to data governance and provide high-quality and comprehensive data sources. For many years, Chinese petroleum enterprises have been committed to building a data governance system with clear classification, reasonable storage, efficient use and sustainable improvement, including security mechanism and data management. Taking PetroChina as an example, in the upstream field of oil and gas, in order to strengthen the enterprise level management of geophysical data in each exploration area, the centralized management of the basin where the mineral rights transfer block is located and the remote backup management needs, the construction of geophysical data cloud data center is carried out, the unified control and management of data and map data are realized, and the comprehensive and complete management of large data in the upstream field of oil and gas is realized; The exploration and production sector will build a dream cloud platform in accordance with the principle of "two unifications and one common use", and gradually integrate multi-level and multi-dimensional data of various disciplines in the upstream of PetroChina through the construction of a data lake to meet the needs of research, production and business management of oil and gas upstream business.
Data governance is the primary link of data system construction. It is suggested to strengthen the data governance work, and realize the availability, easy use and good use of these professional data through the construction of the data governance system, so as to build high-quality, full life cycle data assets in the upstream field of oil enterprises, and provide all-round data support for business research and operation management in the exploration and development field.
Second, build a knowledge map of the field and comprehensively carry out intelligent application exploration. The key to the transformation of oil and gas upstream data into assets is to build the knowledge system of oil and gas data assets through the construction of the knowledge map of exploration and development. With the rapid development of artificial intelligence, especially deep learning and natural language processing technology, knowledge maps show rich application value in assisting intelligent question answering, natural language understanding, big data analysis, intelligent recommendation, Internet of Things device interconnection, interpretable artificial intelligence, etc. Knowledge mapping technology can reduce the threshold for professionals to use knowledge and shorten the time for knowledge retrieval and research; It can quickly discover and mine the value of knowledge, and improve the efficiency of exploration and development decision-making. Therefore, it is of great practical significance to use knowledge mapping related technologies to realize the automation of exploration and development data management, intelligent retrieval, multidimensional analysis and its application in various practices in the oil and gas field.
In order to meet the scientific research and production needs in the oil and gas exploration and development field, we need to design and build a knowledge map of exploration and development. We need to comply with the actual business needs, combine the characteristics of multi discipline and multi discipline collaboration in the upstream field, consider the scientific research business model in the domestic oil and gas field, and build a knowledge map of the field from the perspective of geology, taking basins and oil and gas reservoirs as the main line. Its construction process includes knowledge system classification, ontology model construction, named entity recognition, relationship extraction and knowledge fusion.
It is suggested to speed up the construction of the knowledge map in the whole field of exploration and development, prepare the construction standard of the knowledge map in the field of exploration and development, complete the construction of the knowledge map in the upstream field of oil and gas with the mode of joint construction, and comprehensively carry out the dual drive exploration practice of "data+knowledge" in the upstream field, intelligently solve professional problems, and lead the development of "third-generation artificial intelligence" technology.
Third, R&D exploration and development data middle platform, modular reuse service actual business scenarios. Data middle office refers to the use of new generation information technology to collect, calculate, store and process massive structured and unstructured data, and unify standards to form a big data asset layer, thus providing highly relevant business, unique and reusable efficient data services. The application practice of data science and big data technology realizes the construction and application of data technology capability and data assets through the data middle platform.
Build data services through the data center to realize the interoperability and sharing reuse of applications and data. Split, decouple and encapsulate the data in the application system into services, form new operation management logic, break the information island, realize application integration, function modularization and service-oriented agile development, realize business data and data business, and meet the requirements of collaborative research and business application.
In the exploration and development field, there are complex dependencies and associations in both disciplines and data. The same type of data will support services in different business scenarios. Therefore, it is recommended to comprehensively promote the construction of exploration and development data middle office, unify data, unify identification, refine business scenarios, and develop standard application modules, Realize the reuse and support of different business scenarios, and pave the way for mining the business value of data assets.
In the future, through the in-depth application of data science and big data technology, we will build a digital world with full perception, full link, full scene, and full intelligence in the upstream field of oil and gas, and then optimize and reconstruct the business in the physical world, realize the comprehensive innovation and remodeling of the traditional management model, business model, and business model, fully build a digital ecosystem, and achieve digital innovation, Improve the competitiveness of enterprises. (Zhou Xiangguang Information Technology Center of China Petroleum Exploration and Development Research Institute)
The digital upgrading of international oil companies shows their talents
Shell: Digital Twin Technology
Shell has always been optimistic about the prospect of digital twin technology. In September last year, Akselos deployed structured digital twin technology for Shell's Bonga Main FPSO in Nigeria. Last October, Shell signed a cooperation agreement with Aveva to support the application of digital twin technology in the life cycle of managed assets by building an engineering data warehouse.
Eni: supercomputing and algorithms
Eni regards big data processing ability as a competitive advantage and launches a powerful industrial computer - HPC5 supercomputer. In addition to strengthening industrial data processing capacity, Eni also focuses on artificial intelligence, human-computer interaction, industrial Internet of Things, robot and additive manufacturing, and blockchain technology.
ADNOC: Panoramic Digital Command Center
ADNOC established Panorama, a panoramic digital command center, which collected real-time information from 14 professional subsidiaries and joint ventures, predicted a series of operational scenarios through intelligent analysis models, artificial intelligence and big data, and gave effective operational insights and suggestions. The continuous investment and construction of digital transformation in the past few years have made ADNOC more flexible and adaptable in the current industry environment.
Schlumberger: machine learning and cloud computing
Schlumberger continues to strengthen the company's machine learning and cloud computing capabilities. Cooperated with AIQ and G42 in Abu Dhabi on the development and deployment of artificial intelligence, machine learning and data solutions in the oil and gas industry; Cooperate with IBM's Red Hat, hoping to combine hybrid cloud computing technology with the oil and gas industry to create a digital platform in the future; He has also cooperated with Google, Microsoft and other companies to support the provision of machine learning services to its customers.
Halliburton: Digital Supply Chain
Halliburton signed strategic agreements with Accenture and Microsoft to help improve its digital capabilities on Microsoft Azure cloud. According to the protocol, Halliburton will improve the database analysis capability by enhancing the real-time platform for remote operation expansion, using machine learning and artificial intelligence, accelerate the application deployment of new technologies, and improve the reliability and security of Halliburton's overall system. Halliburton also cooperated with Accenture to accelerate the transformation of digital supply chain. Use AI analysis to strengthen the visibility and operability of real-time supply chain, so as to improve transparency and make faster decisions. |
