This paper examines the role of digital technologies in reducing carbon emissions in China's energy and power industry. Key technologies such as Internet of Things (IoT), big data, Artificial Intelligence (AI), digital twins, and blockchain are applied in power generation, transmission, and carbon capture and storage (CCS) to enhance efficiency and reduce emissions. In power generation, IoT and sensing technologies optimize combustion processes and support renewable energy integration. Big data and AI improve energy management and predictive maintenance, while digital twins enhance system design and operational efficiency. In power transmission, these technologies enable real-time monitoring, dynamic load adjustments, and optimized dispatch strategies. Blockchain ensures transparency in green electricity trading. In CCS, digital tools improve monitoring and operational efficiency. Despite challenges in technological maturity and data governance, some suggestions have been proposed including standardizing IoT protocols, integrating multi-source data platforms, and optimizing blockchain mechanisms. And Digital technologies can significantly accelerate the industry's transformation towards a low-carbon and intelligent future, supporting carbon neutrality goals.

Thermal power generation is one of the major sources of carbon emissions in the world and in China, mainly from fossil fuel combustion and desulphurization processes. 821 g/kWh of CO₂ is emitted per unit of thermal power generation in 2023 [1], and China has introduced a series of policies to support the green and low-carbon development of the thermal power sector, including the “Action Plan for the Low-Carbon Renewal and Construction of Coal Power Generating Capacity (2024–2027)”, which demonstrates China's determination and action to reduce carbon emissions from the thermal power sector.

Digital technology refers to the general term for a series of technologies that utilize modern information and communication technologies, such as the Internet, big data and artificial intelligence, to collect, process, transmit, store, analyze and apply information. These technologies break the traditional time and space limitations, realize the high-speed flow of information and the optimal allocation of resources, and are innovative and promote economic development.Digital technology plays an important role in helping the global process of combating climate change. Digital technology is deeply integrating with key carbon emission fields such as energy and power, industry, transportation, and construction, enhancing the efficiency of energy and resource use, and realizing the double enhancement of productivity and carbon efficiency, and digitization is becoming an important technological path to achieve carbon neutrality. In this process, digital technologies such as 5G, big data and cloud computing, artificial intelligence, Internet of Things, digital twins, blockchain will play an important role in supporting the realization of Carbon Peak and Carbon Neutrality goal.

Carbon Capture and Storage (CCS) is a critical climate change mitigation technology. Its primary function is to reduce greenhouse gas emissions by capturing carbon dioxide (CO₂) generated during industrial activities and energy production, and then safely storing it underground or in other media. The energy and power industry is the most extensive application area for CCS technology, especially targeting high-emission fixed sources such as coal-fired and gas-fired power plants. So, the application of digital technologies in CCS has been discussed. Figure 1 and Figure 2 show the main infrastructures of energy and power generation and transmission systems and Carbon Capture and Storage (CCS).

(a) thermal power systems

(b) renewable energy systems

(c) power transmission line

Figure1. The main infrastructures of (a) thermal power systems, (b) renewable energy systems and (c) power generation and transmission systems.

This paper discusses the digital technologies in Chinese energy and power generation and transmission systems and Carbon Capture and Storage (CCS) field separately in detail. Finally gives some suggestions on future application of digital technologies to cut carbon emission.

Figure 2. The main infrastructures Carbon Capture and Storage (CCS).

Digital technology has already played an important role in the thermal and clear energy power generation industries, contributing to carbon emission reduction by improving power generation efficiency, energy utilization efficiency, and reducing energy consumption.

2.1. Internet of Things (IoT) and Advanced Sensing Technology

The use of IoT and advanced sensing technologies has significantly reduced carbon emissions from thermal power generation by improving coal combustion efficiency, realizing intelligent control, optimizing energy management and realizing carbon emission monitoring and accounting, among other multifaceted measures. Figure 3 is the main function of IoT and advanced sensing technologies in power generation.

Figure 3. The main function of IoT and advanced sensing technologies in power generation.

• Sensors monitor coal, flame, carbon, oxygen, and airflow in real time to optimize combustion, reduce fuel use and carbon emissions. Oxygen and CO sensors control oxygen levels to cut incomplete combustion. Infrared devices monitor furnace temperatures to boost efficiency [2]. Interconnected equipment enables data acquisition, remote monitoring, and intelligent analysis, enhancing overall efficiency and safety in thermal power plants.
• The IoT-based thermal power carbon emission monitoring system enables remote, unmanned monitoring of real-time carbon emissions and intelligent data analysis with early warnings [3]. For instance, in May 2021, Jiangsu Province's power industry carbon emission measurement system went online, collecting real-time flue gas data from over 2,000 parameters across boilers, turbines, and desulfurization systems to achieve accurate carbon emission measurement [4].
• Using real-time data acquisition and equipment interconnection can optimize renewable energy systems' efficiency and maintenance. Sensors monitor equipment conditions (e.g., wind turbine vibrations, photovoltaic panel temperatures) to predict failures and optimize maintenance cycles, reducing downtime and energy waste [5,6]. Real-time data on grid load and weather enables dynamic energy distribution, enhancing renewable energy integration and reducing fossil fuel reliance. Drones and 5G technologies facilitate intelligent inspections of photovoltaic plants, quickly identifying defects and improving power generation efficiency [7]. For example, the 2024 5G+Smart Distributed Photovoltaic project in Zhaoqing, Guangdong, China, uses 5G and drones to detect photovoltaic module defects, reducing annual carbon emissions by 36.1 tons.

2.2. Big Data Technology and Artificial Intelligence

The use of big data technology and artificial intelligence has significantly reduced carbon emissions from thermal power generation by improving power generation efficiency, precise energy consumption control, whole-process optimization, carbon emission monitoring and early warning, and other aspects. Figure 4 is the main function of big data technology and artificial intelligence in power generation.

Figure 4. The main function of big data technology and artificial intelligence in power generation.

• Big data and AI analyze real-time operations and energy use of thermal power units, building optimization models to adjust parameters and guide flexibility upgrades [8,9]. By examining historical and real-time data, key energy consumption factors are identified to develop effective energy-saving measures. For example, Jingdong Digital Technology Group's “AI+thermal power generation” system uses deep reinforcement learning and neural networks to cut coal use and boost boiler efficiency, reducing energy consumption by 0.5% [10].
• A real-time, dynamic and credible carbon accounting system can be realized through the construction of statistical analysis models and prediction and early warning technology by big data technology [11–13].
• Big data and AI enhance clean energy power generation accuracy and grid management efficiency by integrating multi-source data to build predictive models [14–17]. AI algorithms, combined with meteorological and historical data, create wind/photovoltaic output models that predict power generation, optimize energy storage strategies, and improve grid dispatch, reducing backup thermal power needs. Machine learning analyzes equipment data to predict failures, minimizing downtime. AI models also dynamically adjust photovoltaic panels or wind turbine speeds based on weather data to maximize energy capture. For example, Huawei launched an AI-based renewable energy forecasting solution with China Huadian Corporation and Beijing JiuTian Weather, using advanced meteorological models to optimize energy storage and grid stability. Qinghai Province's Smart Dual-Carbon Big Data Center developed a centralized forecasting model covering 513 stations, reducing wind and solar curtailment and increasing clean energy utilization [18].

2.3. Digital twin technology

The application of digital twin technology in the energy sector significantly reduces carbon emissions through methods such as power generation system optimization design, improving combustion efficiency, predictive maintenance, and minimizing resource waste. Figure 5 shows the main function of digital twin technology in power generation.

Figure 5. The main function of digital twin technology in power generation.

• The use of digital twin technology improves energy security and availability by optimizing combustion efficiency, predictive maintenance, and reducing waste and water wastage, and significantly reduces carbon emissions from thermal power generation [19–23]. Digital twin technology can cover the whole process from coal transportation to waste heat recycling to optimize the thermal power generation process. By establishing digital twin models of combustion, thermal and electrical systems, the performance calculation and optimization model databases are preset in the system, and the analysis results are used as the basis for parameter adjustment in order to improve the productivity of the power plant. [24,25]. For example, a digital twin of a boiler can detect changes in coal, sense environmental conditions and power demand, and determine the optimal operating strategy to maximize thermal efficiency and reduce carbon emissions from the boiler. Digital twin technology is used to simulate pollutant emissions under different abatement measures and optimize abatement strategies to ensure that thermal power plants minimize abatement costs while meeting environmental requirements.
• Digital twin achieves the optimization of physical systems' entire lifecycle through virtual mapping. By simulating photovoltaic power generation under different weather conditions, it optimizes module layout and energy storage configuration, enabling system design optimization and simulation [26–30]. For example, the digital twin system for photovoltaic power stations developed by Qingdao Sarnath uses an intelligent IV curve diagnostic algorithm to detect component-level faults, improving operation and maintenance efficiency by 30% and reducing carbon emissions by about 15%. By constructing digital models of wind turbines, performance deviations are monitored, and maintenance plans are automatically generated, reducing operation and maintenance costs and enabling real-time management [31–34].

The carbon footprint of the transmission sector mainly comes from losses in grid transmission lines. In the transmission segment, the Internet of Things (IoT) and sensors monitor the grid status in real-time, while AI and big data predict grid load and optimize energy dispatch. Digital twin enhances grid stability and predicts failures, and blockchain ensures the transparency and credibility of green electricity transactions. The integration of these technologies promotes the intelligent and green development of the power system.

3.1. Internet of Things (IoT) and Advanced Sensing Technology

Real-time monitoring via sensors and RFID tags identifies inefficient energy links. Data analysis optimizes management, reducing energy consumption and emissions. Smart sensors and edge computing also monitor equipment for defect detection and efficiency improvement. Figure 6 shows the main function of digital twin technology in power generation.

Figure 6. The main function of IoT and advanced sensing technologies in power transmission.

• Through real-time monitoring and data analysis using devices such as sensors and RFID tags, inefficient links in energy consumption can be identified. By analyzing and optimizing management through data, measures can be taken to improve transmission efficiency, significantly reducing energy consumption and carbon emissions. For example, intelligent sensors monitor the operating status of distribution transformers, enabling efficient integrated management such as power inspection, power quality monitoring, and load management, thus reducing electrical losses [35–38]. By monitoring parameters such as current and voltage on transmission lines in real-time, the grid load distribution can be dynamically adjusted, reducing ineffective losses [39–44].
• Smart sensors and edge computing technologies can be used for real-time monitoring, diagnostics, and decision support for equipment’s electrical, mechanical, and operational information, allowing for the timely detection and handling of equipment defects, reducing failures and downtime. For instance, sensors installed on transmission lines can monitor real-time data such as weather conditions, ice coating, micro-wind vibration, conductor temperature, and sag, thus improving transmission efficiency and safety.

3.2. Big Data and Artificial Intelligence Technologies

Big data technology and artificial intelligence (AI) indirectly reduce carbon emissions in terms of improving the safety of transmission equipment, enhancing inspection efficiency, optimizing energy management, and enhancing adaptive capabilities. Figure 7 gives the main function of Big data technology and AI in power transmission.

Figure 7. The main function of Big data technology and AI in power transmission

• Training algorithms on historical data can predict grid load patterns, optimize clean energy dispatch, and reduce coal power reliance [45–51]. Analyzing user electricity data helps adjust tariffs to guide staggered peak consumption and lower grid peak loads. Real-time monitoring of power equipment improves energy efficiency and reduces consumption and emissions. For example, Huawei and Shenzhen Power Supply Bureau's intelligent inspection system uses visual AI to enhance transmission line monitoring, while State Grid Hunan Electric Power Co. and Baidu's system improves power utilization efficiency through big data fusion.
• Electricity big data, with its real-time, high-accuracy, high-resolution, and wide collection range, can be used for accurate carbon emission monitoring and accounting [52,53]. The National Grid Big Data Center has successfully measured monthly carbon emissions for the country, sub-regions, and sub-industries using the Electricity-Carbon Calculation Model [54].
• Grid companies collect carbon emission data via sensors and use big data technology to analyze trends in the green supply chain. They identify weak points for carbon reduction on a micro level and provide decision-making suggestions for green supply chain managers [55].

3.3. Digital twin technology

Digital twin technology synchronizes real-time grid models to optimize load, predict maintenance, aid scheduling, and track emissions, enhancing efficiency and sustainability. Figure 8 shows the function of digital twin in power transmission.

• Digital twin technology creates real-time virtual grid models to optimize load distribution, detect faults, and predict maintenance needs [56–60]. By simulating grid operations under diverse conditions and leveraging historical/real-time data, it identifies anomalies, forecasts equipment degradation, and enables predictive maintenance—reducing downtime, costs, and enhancing reliability.

Figure 8. The main function of digital twin technology in power transmission.

• The technology aids decision-makers in energy scheduling by simulating grid dynamics, improving stability and efficiency. For renewables, it models solar/wind systems under varying environmental conditions to optimize output, resolve performance issues, and refine energy schedules.
• Integrating carbon footprint tracking, digital twins analyze real-time emissions data to pinpoint high-carbon segments, optimize energy use, and reduce emissions, supporting sustainable grid management.

3.4. Blockchain Technology

Blockchain enhances green power trade transparency, enables traceability and smart contracts, with China's pioneering international standard. Figure 9 shows the main function of blockchain in power transmission.

Figure 9. The main function of blockchain in power transmission.

• Blockchain improves green power trade transparency and credibility by addressing traditional issues like complex certification, high costs, and tampering risks [61–68]. In December 2024, China launched its first international standard (IEEE P3224) in 2024, leveraging blockchain’s distributed storage, multi-party consensus, and labeling tech to create a robust certification framework.
• The blockchain-based green electricity consumption certification application achieves the on-chain operation of stages such as green electricity trading declaration, confirmation, and clearing [69]. The entire process data of green electricity trading is recorded on the blockchain, ensuring transparency and immutability of the data.
• Through smart contracts, the automatic execution and efficient settlement of green electricity transactions are realized, enabling data sharing and collaborative management, ensuring the accurate traceability and full lifecycle tracking of every green electricity transaction, activating the vitality of green electricity trading, building a green electricity consumption guarantee mechanism, and enhancing the authority of green electricity consumption certification [70–74]. State Grid Blockchain Technology Company successfully helped the Beijing Winter Olympics and Hangzhou Asian Games achieve 100% green electricity in organizing and hosting the events and green electricity trading.

Digital technologies are profoundly transforming the application of Carbon Capture and Storage (CCS) by enhancing efficiency, optimizing management, and strengthening monitoring capabilities.

4.1. IoT and advanced sensing

IoT and advanced sensing technologies enable real-time CO₂ monitoring, optimize carbon capture efficiency, and predict equipment failures via data analysis. Figure 10 is the main function of IoT and advanced sensing technologies in CCS.

Figure 10. The main function of IoT and advanced sensing technologies in CCS.

• IoT technology enables real-time monitoring and data acquisition of the power system through sensors, smart meters, and communication devices [75–77]. These sensors can monitor the emissions, concentration, and distribution of carbon dioxide, providing accurate data support for carbon capture. Advanced sensing technologies, such as fiber-optic distributed sensing technologies (DTS, DAS, DSS), offer high-precision spatial and temporal resolution data, which are used to evaluate reservoir characteristics, monitor CO₂ bubble distribution, detect microseismic events, and track subsidence [78–81].
• Advanced sensing technologies, such as distributed fiber-optic sensing, can provide high-precision monitoring data, which can be used to track the CO₂ sequestration process, ensuring its safety and effectiveness [82–89]. For example, sensors can monitor the temperature, pressure, and CO₂ concentration in underground reservoirs, allowing for the early detection of potential risks.
• IoT technology can assist in optimizing the operational state of carbon capture equipment by analyzing data to predict equipment failures, thereby improving the reliability and operational efficiency of the equipment [75,90–92]. Through IoT platforms, remote monitoring and management of carbon capture equipment can be achieved, ensuring that the equipment operates in optimal conditions, which in turn enhances carbon capture efficiency.

4.2. Big Data and Artificial Intelligence

The integration of big data and AI technologies in carbon capture and storage (CCS) has not only enhanced the economic viability and efficiency of CCS technologies but also reduced costs, fostering innovation and development in related technologies [93–96]. Figure 11 gives the main function of Big data technology and AI in CCS.

Figure 11. The main function of Big data technology and AI in CCS.

• Big data technologies can optimize the application of CCS technologies, enhancing their economic efficiency. For instance, through big data analysis, the operational mechanisms of CCS technologies under varying load fluctuations can be optimized to improve their flexibility and reliability. Big data can also be used to study the integration, scheduling, and optimization of CCS technologies within the power system, boosting the overall system efficiency.
• AI can improve carbon capture efficiency by adjusting model parameters and reducing energy consumption [95, 97–100]. Research from the University of Surrey has shown that adjusting the system model of coal-fired power plants using AI can reduce energy consumption by 36.3% while maintaining the same capture efficiency and capturing more CO₂. AI technologies can enable real-time monitoring and predictive maintenance of carbon capture systems, allowing for the timely detection of potential issues and adjustments to ensure the stable operation of the system.
• AI plays a crucial role in discovering new carbon capture materials and optimizing existing ones [101–105]. Machine learning algorithms can quickly identify efficient adsorbents and solvents, accelerating the development and application of these materials.
• AI can assist in designing more efficient capture reactors and optimizing operational conditions such as temperature, pressure, and chemical reactions, thereby improving capture efficiency and reducing costs.
• AI has significant advantages in optimizing transportation and storage routes for CO₂, as well as planning dedicated CO₂ storage sites. By optimizing transport routes and storage solutions, transportation costs and risks can be minimized.

The application of digital technologies in the energy and power industry has accelerated the development of smart grids, renewable energy integration, and energy efficiency improvements. However, it also faces multidimensional challenges such as technological maturity, data governance, and system security.

5.1. Internet of Things (IoT) Technology

Main issues:

• Lack of standardized protocols among IoT device manufacturers, resulting in difficulties in data interoperability, device compatibility, and insufficient standardization.
• The massive number of connected devices increases the attack surface, and critical nodes in the power system (e.g., substations) could be vulnerable to malicious attacks, threatening grid stability.
• The wide distribution and complex environments of IoT devices (such as remote wind turbines) increase the cost and reduce the efficiency of traditional manual inspections, and remote maintenance technology is not yet fully mature.

Suggestions:

• Develop a unified communication protocol for IoT devices (e.g., smart grid standards based on IEEE 2030.5) to reduce data silos and promote the construction of standardized protocols.
• Deploy lightweight encryption algorithms (e.g., SM9) on devices and integrate blockchain technology to ensure data immutability, enhancing edge computing and security protections.
• Utilize drones equipped with infrared thermal imaging and AI algorithms to automate equipment status monitoring and fault prediction, and promote AI-powered drone inspections.

5.2. Big Data Technology

Main issues:

• The decentralized nature of power data sources (generation, transmission, and consumption) leads to redundancy, missing data, and inconsistent formats, which affect load forecasting accuracy and make data integration challenging.
• Terabyte-level real-time data requires high-performance storage architectures, and the storage and computing costs are high, making it difficult for small and medium-sized power enterprises to afford cloud computing resources.
• User consumption behavior data may be misused, requiring a balance between data openness and privacy protection.

Suggestions:

• Integrate multi-source data such as SCADA, GIS, and user consumption data, and employ federated learning techniques to enable cross-department data collaboration, establishing a comprehensive data platform.
• Adopt a Hadoop+Spark architecture for tiered storage of cold and hot data, optimizing distributed storage solutions to reduce storage costs.
• Use blockchain smart contracts to control data access permissions, establishing a hierarchical data authorization mechanism to ensure user privacy compliance.

5.3. Artificial Intelligence (AI)

Main Issues:

• Deep learning algorithms lack transparency in fault diagnosis, and the lack of model interpretability reduces trust in AI-based decisions among dispatch operators.
• Renewable energy output forecasting requires large amounts of historical data for training, with a high dependence on labeled data. However, the volatility of wind and solar resources leads to sparse samples.
• AI model training (e.g., large-scale power models) consumes significant amounts of power, creating a conflict between computational energy consumption and green, low-carbon goals.

Suggestions:

• Utilize transfer learning techniques to reuse common model parameters, reducing the need for labeled wind and solar data, and develop lightweight AI models.
• Combine digital twin technology to simulate extreme scenarios, build trustworthy AI frameworks, validate AI decision logic, and generate visualized reports.
• Co-locate AI computing centers near renewable energy plants (e.g., wind farms) to promote green computing infrastructure, achieving a "green power for computing" closed loop.

5.4. Digital Twin Technology

Main Issues:

• There is a delay in the dynamic mapping between physical devices and virtual models; for example, simulation errors in transformer temperature fields impact life expectancy predictions, leading to insufficient model accuracy.
• Digital twin modeling requires the integration of knowledge from multiple disciplines such as electrical engineering, materials science, and fluid mechanics, but cross-disciplinary collaboration is difficult, and the integration of modeling tools is insufficient.
• Some older equipment lacks sensors, resulting in insufficient real-time data support, causing delays in updating twin models.

Suggestions:

• Use AI to dynamically calibrate model parameters (e.g., dynamic error compensation based on LSTM) and develop adaptive twin engines to improve simulation accuracy.
• Build an open API ecosystem, integrating tools like ANSYS and MATLAB, and create cross-domain collaboration platforms to support multi-physics field coupling simulations.
• Deploy wireless vibration sensors and voiceprint recognition devices, promote low-cost sensing technologies, and enhance the data collection capabilities of old equipment.

5.5. Blockchain Technology

Main Issues:

• Low transaction throughput, with existing consensus mechanisms (e.g., PoW) unable to support high-frequency power transactions (e.g., microgrid P2P trading).
• Data silos between carbon trading, green certificate trading, and electricity market chains hinder multi-energy coordination, with poor cross-chain interoperability.
• The decentralized trading model conflicts with traditional power regulatory systems, and there is a lag in policy and regulatory development.

Suggestions:

• Adopt a “main chain + side chain” architecture, where the main chain handles settlement, and the side chain supports high-frequency transactions, optimizing the layered consensus mechanism.
• Collaborate with grid companies and renewable energy producers to create cross-chain protocols, building an energy blockchain alliance chain to enable carbon, electricity, and certificate data interoperability.
• Launch distributed energy trading pilot projects in free trade zones, promote sandbox regulatory trials, and explore compliance-based smart contract templates.
• Through technological integration and innovation, policy and standards collaboration, and the construction of cross-industry ecosystems, digital technologies will accelerate the transformation of the energy and power industry toward a "safe, efficient, low-carbon, and intelligent" direction, providing core support for the development of new power systems.

Digital technologies (IoT, big data, AI, digital twins, blockchain) facilitate the low-carbon transition of the power sector by optimizing energy generation, transmission, and carbon capture and storage (CCS). IoT enables real-time equipment and emission monitoring, enhancing combustion efficiency; AI and big data optimize grid dispatch and fault prediction; digital twins simulate system operations to reduce maintenance costs; blockchain ensures transparent and trustworthy green power transactions. In CCS, high-precision sensing and AI improve carbon sequestration monitoring and material innovation. Despite challenges like data governance, security, and technical maturity, solutions such as standardized protocols, lightweight encryption, and cross-chain collaboration can drive integration. These technologies support the evolution of energy systems toward safety, efficiency, and low-carbon operations, providing core support for carbon neutrality goals. Future efforts should focus on interdisciplinary collaboration, policy alignment, and scalable green computing to accelerate the development of smart, sustainable power systems.

Conceptualization, Huichun Xie and Ying Liu; methodology, Huichun Xie; formal analysis, Huichun Xie; investigation, Huichun Xie and Ying Liu; resources, Huichun Xie and Ying Liu; writing—original draft preparation, Huichun Xie and Ying Liu; writing—review and editing, Huichun Xie; visualization, Ying Liu.

The authors declare no conflicts of interest.