Global Visibility | Wu Moueyuan, Vice President of China National Petroleum Corporation Research Institute: How Digital Intelligence Technology Reshapes Energy Efficiency

How AI and Intelligent Digital Technologies Are Solving Supply Security and Transformation Challenges in the Oil and Gas Industry

[Global Network Finance Report, Reporter Chen Chao] 2026 marks the beginning of the 14th Five-Year Plan, with clear development goals for the energy sector. By 2030, the comprehensive energy production capacity should reach 5.8 billion tons of standard coal, and carbon emissions per unit of GDP should decrease by 17% compared to 2025. These hard targets define the dual core missions for the traditional oil and gas industry, which serves as the “ballast” of national energy security: ensuring a stable energy supply and safeguarding the bottom line of oil and gas supply, while also addressing deep-seated industry contradictions to shift from a “fuel-based” to a “raw material and material-based” value chain, achieving green, low-carbon, high-quality development.

Currently, the traditional oil and gas industry faces dual structural reforms: on the resource side, the industry has entered the “two deep and one non” development era, with high-quality, easily developed conventional resources continuously declining, and increasing difficulty and costs in exploration and development, intensifying conflicts between supply security and resource endowments. On the demand side, global refined oil demand has plateaued, high-end chemical new materials demand continues to surge, and traditional fuel refineries face structural contradictions of “more oil, less refined products.” The industry must accelerate its transition from “fuel” to “raw materials,” cultivating new growth curves.

(吴谋远, Vice President of China Petroleum Group Economic and Technical Research Institute, provided by Sinopec Economics and Research Institute)

Traditional production management models are no longer sufficient to meet industry transformation needs. Four major challenges—supply security, cost reduction, safety, and transformation—must be addressed. Where is the breakthrough? In a recent exclusive interview with Global Network, Wu Mouyuan, Vice President of China Petroleum Group Economic and Technical Research Institute, provided a systematic answer: AI, big data, and other intelligent digital technologies have evolved from auxiliary tools into strategic engines that reshape the core competitiveness of the traditional energy industry. They are not only key to strengthening supply security but also critical to transforming from “fuel” to “raw material” and reconstructing the value chain. Wu believes that AI and big data are not merely empowerment tools but the “brain” and “nerves” of a new energy system.

Intelligent Digital Technologies: Breaking Bottlenecks in Exploration and Development, Strengthening Supply Security

The primary mission of the traditional oil and gas industry is supply security. The transformation from “fuel” to “raw materials” fundamentally depends on efficient upstream resource development. Wu points out that the full-chain intelligent upgrade driven by digital technologies is the core path to solving both “resource development bottlenecks” and “downstream raw material transformation needs,” promoting the entire process of exploration, development, and production from mechanization to high automation and autonomous decision-making.

In exploration and development, AI algorithms deeply analyze geological big data, significantly improving exploration success rates. The interpretation of seismic data and reservoir identification, which traditionally took months, can now be compressed to days, directly solving the “two deep and one non” exploration and development challenges. More importantly, digital technologies enable precise identification of crude oil molecular components, allowing early prediction of aromatic hydrocarbons and olefins content, providing a precise raw material basis for flexible production—“oil when suitable, olefins when suitable, aromatics when suitable”—supporting the source-to-transformative shift from fuel to raw material.

In production operations, integrating AI with IoT, automation, and robotics enables unmanned inspections and predictive maintenance at production sites, significantly reducing lifecycle operation costs. This approach has been validated by leading domestic and international companies: domestically, CNPC’s Changqing Oilfield has built the largest oil and gas IoT system in China, achieving full digital coverage of over 100,000 wells and over 2,500 facilities with unmanned operation, ensuring long-term stable production and efficiency; internationally, Norway’s Statoil (now Equinor) has used AI-based geological modeling to greatly improve deepwater reservoir identification accuracy, shortening exploration cycles by 70%; ExxonMobil in the Permian Basin has optimized fracturing schemes with AI, significantly increasing well productivity and development efficiency.

Intelligent Digital Technologies: Building Proactive Defense for Traditional Energy Safety Production

Safety is the lifeline of the traditional energy industry and a bottom-line guarantee for the transformation from “fuel” to “raw materials.” The processes of refining and high-end chemical raw material production are more complex and risky, demanding higher safety controls. Historically, safety management in the industry has been reactive and passive, with high-risk scenarios such as oil and gas fields, refineries, and pipelines facing issues like delayed hazard detection and imprecise risk warnings.

Wu states that digital technologies are fundamentally reconstructing the safety management system of the traditional energy industry, enabling the reinforcement of safety resilience systems and achieving a leap from “reactive handling” to “pre-emptive risk avoidance.” Digital tools can establish real-time, full-lifecycle proactive safety defense and emergency response systems for energy infrastructure, utilizing edge computing and computer vision to enable real-time risk identification and early warning in high-risk scenarios like refining units, pipelines, and storage facilities, thus fortifying the safety bottom line for the transformation from fuel to raw material.

This system has been scaled globally. Domestically, oil and gas refining companies widely adopt AI-based intelligent inspections and full-process digital twin early warning systems, greatly reducing safety incidents. Internationally, Pacific Gas and Electric (PG&E) in the US uses AI risk warning systems to predict wildfires and line faults 72 hours in advance, significantly reducing fault occurrence; Spain’s power grid employs AI drone inspections for nationwide transmission line coverage; Korea Electric Power Corporation (KEPCO) has implemented intelligent substation maintenance systems, greatly lowering equipment failure rates.

Intelligent Digital Technologies: Empowering Green Transformation and Driving Value Leap from “Fuel to Raw Material”

Under the dual carbon goals, traditional energy companies focus on two core directions: integrating renewable energy for green, low-carbon development, and fundamentally reconstructing the value chain by shifting from “fuel” to “raw materials.” Wu believes that digital technologies are the core support for traditional oil and gas companies transforming into integrated energy service providers. Using AI deep learning models to build wind and solar power forecasting systems can accurately capture resource variation patterns, and combined with smart dispatching systems, achieve millisecond-level power balancing, providing stable green electricity for refineries and chemical parks, and enabling low-carbon raw material production throughout their lifecycle.

Domestic examples include Jilin Oilfield, which has built a “wind-solar-geothermal” multi-energy complementary system through digitalization, successfully producing “zero-carbon crude oil,” providing resource support for downstream low-carbon raw materials. Globally, Texas’s ERCOT grid uses AI-based dispatching systems to significantly improve wind and solar power prediction accuracy, avoiding large-scale blackouts during peak summer periods in 2025; the EU’s ENTSO-E cross-border AI dispatching system reduces inter-regional peak response times from hours to minutes.

The Ministry of Industry and Information Technology’s “Work Plan for Stabilizing Growth in the Petrochemical and Chemical Industries (2025-2026)” explicitly promotes the transformation of refining from “fuel-based” to “chemical-based,” implementing “AI + Petrochemical” initiatives. “From fuel to raw materials” has become a core theme of global refining value redefinition, with digital technologies serving as the key enablers. Wu emphasizes that the full-chain intelligent upgrade driven by digital tech is not limited to exploration and development but extends throughout refining, processing, and product structure optimization. Through precise control and deep AI algorithm optimization, traditional refineries are fundamentally shifting from producing fuels like gasoline and diesel to high-end chemical materials and specialty chemicals.

Globally, “from fuel to raw materials” has become a common strategic direction for major energy companies. Saudi Aramco is collaborating with Honeywell to develop next-generation direct crude-to-chemicals technology, aiming to convert each barrel of oil directly into high-end chemicals like light olefins, maximizing oil value; Shell’s Singapore refinery uses digital twin and AI for full-process optimization, significantly increasing chemical product share; South Korea’s SK Energy and Japan’s JX Nippon Oil & Energy are also leveraging AI-based systems to promote refinery transformation toward chemical production.

Future Directions: Four Major Breakthroughs to Secure Long-term Competitiveness of Traditional Energy

Looking ahead to the 14th Five-Year Plan and beyond, Wu proposes that deep integration of AI with traditional energy will achieve systematic breakthroughs in four key areas, driving higher-level digital transformation and providing sustained technological support for the profound shift from “fuel” to “raw materials.” This is also the core battleground for traditional energy industry to seize future industry high ground.

1. Evolution of energy industry large models from “single-task general” to “all-scenario experts.” The key to reshaping the industry lies in developing vertical large models tailored to energy. Pretraining on trillions of energy-specific texts will endow these models with cross-disciplinary reasoning and complex system optimization capabilities, enabling them to handle not only basic visual recognition and production forecasting but also core tasks like exploration planning, refining process optimization, new material formulation, and market decision-making—achieving a leap from “perceptual intelligence” to “cognitive intelligence.” This is a global industry trend; giants like Chevron and Shell are already developing industry-specific large models focused on refining and new materials.

2. Breakthroughs in “Physics-Informed Neural Networks (PINN)” integrating physical laws. Deep fusion of AI with physical characteristics of energy systems is essential for improving computational efficiency and accuracy. Future models will embed physical constraints—such as reservoir flow mechanics, refining thermodynamics, and pipeline fluid equations—directly into neural networks, enabling real-time, precise simulation of complex reactions, overcoming R&D bottlenecks for high-end chemicals, and shortening development cycles for high-value raw materials, thus providing a technical foundation for the deep transformation from fuel to raw material.

3. Fully automated energy market intelligent bidding and trading systems. As market-oriented reforms deepen across oil, gas, electricity, carbon, and chemical markets, AI-driven automated trading will become key to enhancing competitiveness. Deep insights into raw material prices, trading behaviors, and network constraints will support second-by-second market interactions involving millions of participants, helping enterprises maximize value in procurement, sales, and carbon asset trading. Globally, leading companies in the US and EU have adopted AI automation trading systems, becoming core competitive advantages.

4. Distributed intelligent coordination and self-organizing networks under the energy internet. Future digital technologies will enable billions of distributed energy nodes and chemical production units to achieve self-perception, self-decision, and self-coordination. Using edge computing and blockchain, traditional energy networks will evolve into self-healing, self-organizing systems, facilitating coordinated operation among refineries, new energy sources, energy storage, and downstream chemical enterprises, further enhancing raw material stability, low-carbon operation, and economic efficiency.

In today’s era of profound global energy transformation and fundamental industry demand restructuring, the traditional oil and gas industry is undergoing dual leaps: upstream resource development is overcoming “two deep and one non” challenges through digital intelligence, safeguarding energy supply; downstream refining and chemical sectors are driven by digital intelligence to reconstruct value from “fuel” to “raw materials,” opening new long-term development space. As Wu notes, the reshaping of the traditional energy industry by digital technologies is never just about point tools but a systemic transformation across the entire industry chain and lifecycle. Digital intelligence is pushing the energy system from “physical connections” to “logical collaboration,” enabling real-time interaction between digital twins and physical systems, integrating energy flow, information flow, and value flow, providing comprehensive solutions to core industry pain points.