In order to meet the development requirements of PetroChina’s upstream business in the digital intelligent age, Petro- China Exploration & Production Company has carried out top-level design for informatization of the upstream business. This paper describes the current development phase of the new systems in the context of international information development trends and the information technology requirements for development of the upstream business. The overarching goal, and the company’s development vision for the future, is to build fully intelligent oil and gas fields. To achieve this, eight targets are proposed: intelligent data eco-environment, first-class intelligent platform, intelligent collaboration in research, intelligent optimization of program decision-making, intelligent control of production processes, intelligent command of production and operation, lean and efficient operational management, and intelligent control of safety and environmental protection. This paper describes the overall architecture of upstream business informatization, including business architecture, data architecture, technical architecture, application architecture, and network security architecture, as well as the corresponding work priorities and support systems, etc. This represents a full blueprint for informatization of PetroChina’s upstream business.

The E&P Dream Cloud platform is an information sharing platform for the upstream business of PetroChina. One of the core, and most widely used, applications of upstream informatization is the Oil-Gas-Water Well Production Data Management System, known as the A2 system. However, the system contains massive volumes of data with the result that querying, retrieval, and data analysis are still somewhat inefficient. The openness and scalability of the E&P Dream Cloud platform has allowed the A2 system to be upgraded and transformed into a cloud system to enable integrated collaborative management of the collection, processing, storage, statistical analysis, reporting, and publishing of production data from all of the company’s oil, gas, and water wells. With the goal of lean oil and gas production management, data management tracking and evaluation standards have been introduced and application research work has been carried out. This includes production data tracking and evaluation, productivity construction evaluation, and analysis of stripper wells and long-term shutdown wells, all of which provide data support for strategic planning research.

Research on oil and gas reservoirs involves many different disciplines and multiple data types, and requires input of research results from many fields. Multi-disciplinary collaborative research has therefore become an essential and growing trend. In order to provide researchers with an integrated, collaborative and sharing working environment for oil and gas reservoir research, project studio for different research objectives has been constructed, which is on the basis of the E&P Dream Cloud platform and basin-level regional Data Lake structure. It incorporates centralized and standardized data organization, fast querying, integrated application of business software, online auxiliary analysis tools, and whole process management of oil and gas reservoir research projects. The results have been applied to research and risk management of exploration projects in the Ordos Basin. The construction of collaborative research environment, based on E&P Dream Cloud, achieves effective sharing of data, software and research results, and facilitates collaborative working in oil and gas reservoir research, providing supports for the exploration and development business of oil and gas fields and greatly improving the quality and efficiency of scientific research work.

Sub-lacustrine fans in the highstand system tract in the first member of the Liushagang Formation (the Liu-1 member) in B sub sag of the Weixinan sag are an important field for lithologic reservoir exploration. However, the characteristics, classification, distribution, genetic mechanisms, and development patterns of sub-lacustrine fans are not well understood, which constrains the exploration process and achievements. In this study, drilling, core, well logging, laboratory analysis and 3D seismic data are used to systematically describe the sub-lacustrine fans in the highstand system tract in the Liu-1 member in B sub-sag. The results show that: (1) The sub-lacustrine fans in the highstand system tract in the Liu-1 member in B sub-sag can be divided into two major types and four sub-types: mass-flow sub-lacustrine fans in the early highstand stage (which can be further divided into incised-channel center type and incised-channel flank type), and debris-flow sub-lacustrine fans in the late highstand stage (which can also be further divided into non-channelized sand-rich type and non-channelized mud-rich type). (2) Different types of sub-lacustrine fans all present different characteristics in lithology and in well logging and seismic data. (3) There are two development models, which are jointly controlled by lake level changes, sediments source systems, and transport routes. The first depositional model is meandering river delta - sedimentary flexure slope break - mud-rich sub-lacustrine fan with the sediments source co-axial with the sub-sag to the west. The second depositional model is fan delta - fault transition slope break - sand-rich sub-lacustrine fan with the sediments source to the northwest of the sub-sag. (4) Sub-lacustrine fan sand bodies of the incised- channel center mass-flow type and the non-channelized sand-rich debris-flow type, supplied by fan delta sediments from the northwest source, have shallow burial depths and good reservoir properties. The sand bodies form a self-generating, self-storage, and self-sealing hydrocarbon accumulation assemblage with the lacustrine mudstones in the Liu-1 member—an assemblage which has great exploration prospect.

In this paper, continental shale gas reservoirs in the Da’anzhai member of the Jurassic Ziliujing Formation in the Yuanba and Fuling areas in the northeastern Sichuan Basin is studied to investigate the geological characteristics and exploration and development potential. Core analysis and lab test were conducted, including core observation, thin section identification, X-ray diffraction, organic carbon content analysis, Scanning Electron Microscope, high-pressure mercury injection and nitrogen adsorption, to determine the sedimentary environment and lithology, organic geochemical characteristics, reservoir space and gas-bearing properties, mineral composition and fluid properties of continental shale gas reservoir. The results show that organic- rich shales are developed in shallow to semi-deep lacustrine facies in the Da’anzhai member. The lithology is mainly shale intercalated with stripped or laminated sandstones and shell limestones. The organic matter type is Type Ⅱ2 or type Ⅲ. TOC is 0.33%?3.78%. Ro is 1.11%?1.82%, indicating a mature to high mature stage. The reservoir space in the shale is mainly mineral pores. The porosity ranges from 1.5% to 6.7%, with micro-pores less than 2nm and medium pores of 2?50 nm being the most prevalent. The gas-bearing properties are good (average gas content is 1.49 m3/t). The shale has good fracability and the average content of brittle minerals is 51%. A total of 12 wells have obtained high-production oil and gas flows, showing good exploration and development potential. Comparative study of the differences between the characteristics of continental shale and marine shale indicates that favorable sedimentary facies is an important basis in enrichment and high production of continental shale gas in the Da’anzhai member. Shale gas accumulation is also controlled by rock association and lithologic composition, organic carbon content, physical properties, fluid pressure, and the development degree of micro fractures. These understandings provide reference for selection of the most favorable exploration and development target formations of Jurassic continental shale gas in the Sichuan Basin.

The E&P Dream Cloud platform developed by PetroChina contains a unified Data Lake, a unified technology platform, and a general application environment. It supports the creation of a cloud-native exploration research environment which has been implemented and applied in oil and gas geophysical exploration companies in China. The key factors for improving efficiency and results quality in geophysical prospecting are to integrate seismic acquisition, processing, and interpretation, to integrate ‘front’ and ‘rear’ operations (on-site and at base), and to integrate companies running and administering projects with contractors working on their projects. Relying on the Data Lake resources of the individual oil and gas field companies within Dream Cloud and the customizable development capability of the E&P Dream Cloud platform, this paper studies and explores the characteristic mode of integration of seismic data processing and interpretation within the system, and constructs a work and research environment for the integration of front and rear operations, as well as the integration of project companies and contractors. This research advances the ongoing transformation from traditional seismic data processing and interpretation, builds project scenarios for seismic processing and interpretation integration, takes full advantage of the “super project organization” capability in the E&P Dream Cloud, and achieves integration between the project company and its contractors throughout the process of seismic processing and interpretation. It also achieves efficient collaboration and software resource sharing between front and rear, and provides effective support for improving the core competitiveness of enterprises.

In accordance with the government’s policy to accelerate informatization in China, PetroChina has adopted a strategy of “sharing PetroChina”. Since the launch of the E&P Dream Cloud platform, informatization in PetroChina’s oilfields has progressed rapidly. The Tarim Oilfield Company has constructed a regional Data Lake for the oilfield, based on the E&P Dream Cloud platform, to improve the efficiency and management of collaborative research, focusing on the upstream petroleum exploration business. The system meets the requirements for a unified technology platform and unified database, creates a collaborative research environment for risk exploration, and facilitates collaborative management of exploration, development, operations, and production in the oilfield. The trap management module in the system was developed to enable online reporting of traps, fast querying of existing results, online browsing, and three-level management and review. Drawing on successful early application of the system, this paper offers ideas and specific suggestions for the future development of the E&P Dream Cloud. Six aspects are discussed: real-time control of production performance, whole life cycle management of projects, all-round data tracking and querying, intelligent identification of research objects, modular integration of business applications, and integrated display of multi-disciplinary achievements.

Construction of the E&P Dream Cloud platform focuses on a unified database, unified technology platform, and general applications. Firstly, it has achieved a number of general objectives, such as creating a collaborative research work environment for the exploration business and for well deployment demonstration and decision-making, and setting up a demonstration application environment for cloud applications to support collaborative research projects in regional companies. PetroChina Southwest Oil & Gasfield Company (Southwest Oil & Gasfield Company) has carried out an in-depth review of the research requirements for risk exploration in the Sichuan Basin with particular reference to the capabilities of the E&P Dream Cloud platform. As a result, many aspects of risk exploration research in the Sichuan Basin have now been carried out on Dream Cloud. These include: data organization and management, regional geological settings overviews, and research on structure, sedimentary patterns, source rock conditions, favorable plays and prospect targeting, well location design and deployment, etc. A collaborative research environment of “platform + project + business” has been constructed, and a new approach to scientific research and well deployment implementation explored and created in risk exploration based on the E&P Dream Cloud, which effectively supports research and decision-making in risk exploration in the Sichuan Basin and has already achieved positive results in practice.

The Dagang Oilfield has entered the middle and late stages of exploration and development. Scientific research work is encountering problems with the collation of huge volumes of project data, diversity of software applications, and difficulty in sharing results between independent teams. To improve the efficiency and quality of research, a unified platform is urgently needed for project progress control, task allocation, and results archiving to impose order on this project. Software integration based on the E&P Dream Cloud platform integrates business software and common tools such as GeoEast, GeoMap, Resform, Forward and wellbore integration tools in the cloud. It creates a research environment based on core business scenarios such as seismic interpretation of structures, research on characteristics of sand body distribution, prediction of reservoir lateral distribution, research on hydrocarbon accumulation, well logging evaluation, oil and gas bearing property evaluation, etc. Timely and accurate transmission of professional and production data to the Data Lake is enabled, as well as whole process operation and management of projects, quality control, and sharing of applications and results. Online collaborative research work has been carried out, including seismic structure interpretation and analysis, distribution characteristics of reservoirs and sand bodies, and controlling factors of hydrocarbon accumulation. This supports the transformation from traditional “structural oil exploration” to “lithologic oil exploration in main sand body belt” in the study area. The effect in practice has been remarkable, and has played an important role in demonstrating the capabilities of the E&P Dream Cloud platform.

The petroleum industry has entered a new era of rapid digital development and intelligent operations. Maximizing the capabilities of the new technologies is an urgent requirement in every sphere, so the question of how to achieve digitalization and intellectualization of information and management systems has become an important topic for upstream oil and gas enterprises to study and explore. Influenced by the latest development trends in global information technologies, and inspired by international best practice, PetroChina’s upstream business segment has formulated a blueprint of constructing a platform for upstream business information and application sharing—the E&P Dream Cloud platform (Dream Cloud). Dream Cloud adopts a cloud computing and microservice architecture design, merging enterprise data governance concepts with technology systems, and integrating big data, artificial intelligence, and other advanced technologies. The results are a “platform + capability + application” ecology of Dream Cloud and the establishment of an open and sharing environment of “data + technology + application” for the upstream business. The system has already enhanced the intelligence sharing capability of the upstream business, providing agile digital and intelligent services for upstream business applications. The platform and its applications use Devops, agile development, and other technologies to achieve efficient and integrated development which supports data interconnection, technology interoperability, and business collaboration across the upstream business. As an intelligent sharing platform for PetroChina’s upstream business, Dream Cloud has opened up a sustainable development road for the implementation of the company’s overall “sharing Petro- China” strategy and provided support for the digital transformation and intelligent development of the upstream business.

CNPC has been continuously developing its information infrastructure for more than 30 years, going through three stages: decentralized construction, centralized construction, and integrated applications. Now the company is moving forward to a new stage of collaboration and sharing—a “sharing CNPC”. There are “three-multiple” issues in CNPC’s upstream business: multiple platforms, multiple databases, and multiple isolated applications. In order to meet the requirements of digital transformation and development, combining with actual business needs and the general development trend of information technologies, CNPC has created the E&P Dream Cloud and Data Lake technologies. This development required systematic study of the data resource systems of the upstream business and deep analysis of data aggregation, storage, and applications. The Data Interlinked Lakes scheme has been designed and constructed based on the unified Data Lake that is part of Dream Cloud v1.0. The structure of this scheme includes a group-level main data lake and individual data lakes for each regional oil and gas field company, all of which are logically unified and interconnected. Pilot verification of the system achieved the expected results, providing strong support for the construction of a new data eco-environment: Dream Cloud v2.0.

A new seismic data interpretation mode has been proposed, which is based on the E&P Dream Cloud and deep learning methods. This new approach explores the combination of cloud computing and artificial intelligence technology in seismic data interpretation, taking advantage of their collaborative capabilities. It is based on practical R&D experience with deep-learning seismic interpretation software, which includes: (1) Data Lake and Computing Center of the E&P Dream Cloud provides a large volume of high-quality labeled data and a highly elastic computing platform on which to run the deep learning algorithm. (2) Deep-learning technology and traditional geophysical methods complement each other to establish a new process for seismic and geological interpretation. (3) The deep-learning module in the E&P Dream Cloud platform provides effective information support from seismic results for business decision-making within the E&P Dream Cloud. This mode considerably reduces the number of manual operations required and achieves higher prediction accuracy, laying a solid foundation for high-efficiency and high-precision oil and gas exploration. Experience in practice with real data has already produced encouraging results.

Geophysical logging is still the most important technical means used to identify the physical properties of subsurface strata and to decide whether they are hydrocarbon-bearing. Using the technologies of big data and artificial intelligence, based on the E&P Dream Cloud intelligent data analysis platform, this paper presents a new approach to intelligent well logging interpretation. Features of log data are extracted on the basis of data collection and splicing. Expert knowledge and experience are combined to build model. Machine learning and model training are carried out, and model testing and iteration are completed. For this study, intelligent logging interpretation technology was verified by confirming sandstone and mudstone differentiation (with a cut-off value of 0.4 based on 60 wells) and oil and water layer differentiation (with a cut-off value of 0.5 based on 47 wells). The interpretation results show that the intelligent system takes about 2-4 hours to distinguish between sandstone and mudstone. The accuracy rate of mudstone and sandstone identification is 94% and 90% respectively with an overall accuracy rate of 92%. It takes about 4-6 hours to identify oil and water layers. The accuracy rate of oil and water layer identification is 93% and 70% respectively with an overall accuracy rate of 86%. This technology can effectively increase the automation degree and the coincidence rate of multi-well logging interpretation. The application of big data analysis technology provides a new motive force and good prospects for the development of well logging interpretation technology.

Internet technology is very much coming into its heyday and is now entering the era of “Internet plus”. Especially, cloud services and cloud applications are supporting rapid informatization in every industry. The petroleum industry, and other traditional energy industries, have also been carrying out information system construction. However, there are major challenges facing the petroleum industry in how to achieve migration to cloud-based computing—essential for the future to take advantage of the cloud’s ability to support rapid development of business software and applications. Petroleum research has its own specific requirements and a need to rapidly acquire business software and other technology for information system construction such as big data, artificial intelligence, etc. The Dream Cloud App Store solves this problem and fills the requirement for an application sharing platform. The App Store gathers together professional petroleum software, artificial intelligence, and big data technology. Based on E&P Dream Cloud and relying on the Data Lake and PaaS platform within the Dream Cloud environment, the App Store provides an end-to-end utilization interaction (UI) platform for the petroleum industry, effectively supporting application sharing and construction in the oil and gas industry, accelerating rapid informatization and promoting an opening, collaborative, sharing and intelligent eco-environment within the upstream business.