21世纪经济报道记者雷椰 李德尚玉 实习生刘诗媛 北京报道 全球气候治理迈入"落实《巴黎协定》十年"的关键节点，人工智能等前沿技术驱动的算力产业在持续扩 张，同时也带来了巨量的能源需求与碳排放压力。而有关AI、能源议题也成为《联合国气候变化框架 公约》第三十次缔约方大会（COP30）的讨论焦点。 在COP30召开期间，在巴西COP30中国角的现场，BCI Group创始人、CEO居静和他的团队向国际展示 了其绿色算力基础设施样本。居静在接受21世纪经济报道专访时表示，破解AI算力与能源消耗的矛 盾，不能仅靠单点技术，而必须通过"垂直一体化"的生态重构，在源头实现绿色电力直供与全链条协 同，才能真正支撑算力产业的可持续发展。 《21世纪》：今年COP30大会在巴西举行，作为数次参加COP大会的中国企业，从迪拜、阿塞拜疆到 今年的巴西贝伦，一路走来，你和团队有什么特别的感受？对本次大会有怎样的期待？又有哪些收获？ 居静：作为算力基础设施企业，对于今年COP30议题，我们期待气候融资的落实，特别是整合地方政 府、企业与民间社会力量，共同落实能源转型相关内容。具体有几方面的期待：一是可再生能源规模化 增长背景下，气候 ...

Core Viewpoint - The article emphasizes the growing demand for green electricity in China's computing power centers, highlighting the need for collaboration between energy and computing sectors to achieve sustainability goals by 2030 [1][8]. Green Electricity Consumption - Current green electricity consumption in domestic computing power centers is low, with significant growth potential to reach the national average consumption level of 40% by 2030 [1][8]. - Bloomberg New Energy Finance predicts that the demand for green electricity in domestic computing power centers will reach 72 TWh by 2026 [1][8]. Energy Infrastructure Development - The Dazhong Park of Qineng Technology features a 300 MW wind power, 200 MW solar power, and a 50 MW/100 MWh energy storage station, with a planned IT capacity of 1000 MW [3]. - The park aims to achieve grid connection for green electricity by 2024, reflecting the shift from real estate-driven growth to energy-driven development in the computing industry [3][4]. Policy and Market Dynamics - The National Development and Reform Commission and the National Energy Administration have issued guidelines to promote the synergy between renewable energy and computing facilities [3]. - The core customers for green electricity consumption include steel, chemical, and computing centers, with the latter showing stable electricity usage and significant growth potential [4]. Global and Domestic Energy Trends - The International Energy Agency reports that global data center electricity consumption is expected to rise from 415 TWh in 2024 to 945 TWh by 2030, with China and the U.S. contributing nearly 80% of the increase [6]. - By 2030, China's data center electricity load is projected to reach 100 million kW, with annual consumption rising to between 400 TWh and 600 TWh, increasing its share of national electricity consumption from under 2% to 6% [6]. Green Electricity Connection Models - There are three main methods for green electricity consumption: direct connection, green microgrids, and power purchase agreements [9]. - The direct connection model allows renewable energy sources to supply electricity directly to users, bypassing the public grid, which is encouraged by recent government policies [9]. Regional Development Initiatives - Various provinces are launching plans for green electricity parks, with Shanxi Province planning to establish 13 pilot parks across multiple cities [10]. - The selection of sites for green electricity parks is critical, with proximity to renewable energy sources being a key factor for economic viability [12]. Cost Management Strategies - The "green electricity direct connection" model faces high initial investment costs, requiring infrastructure such as solar and wind farms, substations, and backup power systems [16]. - Companies are focusing on lifecycle cost management and vertical integration to optimize resource development and reduce costs [16][17].

Core Insights - The article discusses the growing demand for green electricity in the context of the expanding computing power industry, emphasizing the need for "power-computing synergy" as a trend in energy consumption and infrastructure development [2][3][4]. Group 1: Industry Trends - The computing power industry is shifting from a real estate-driven model to one that requires energy iteration support, making "power-computing synergy" a necessary trend [2]. - The National Development and Reform Commission and the National Energy Administration have issued guidelines to promote the coordinated planning and operation of renewable energy and computing facilities [2]. - The global data center electricity consumption is projected to reach approximately 945 terawatt-hours by 2030, with China and the U.S. contributing nearly 80% of the new electricity demand [4]. Group 2: Green Electricity Demand - The core customers for green electricity consumption include steel, chemical, and computing centers, with the latter showing significant growth potential and stable electricity demand [3]. - Data centers have a high and rapidly growing electricity consumption, making them a key focus for green electricity policies [3][4]. - By 2030, it is expected that the electricity load of data centers in China will reach 100 million kilowatts, with annual consumption rising to between 400 and 600 terawatt-hours [4]. Group 3: Supply Side Developments - The rapid growth of installed renewable energy capacity in China has led to a decrease in the average utilization hours of power generation equipment, necessitating new applications for green electricity [5]. - As of September 2025, solar power capacity reached 1.13 billion kilowatts, a year-on-year increase of 45.7%, while wind power capacity reached 580 million kilowatts, up 21.3% [5]. - The current low consumption ratio of green electricity in computing centers indicates significant room for growth to meet the national target of 40% by 2030 [5]. Group 4: Green Electricity Consumption Models - There are three main methods for green electricity consumption: direct connection, green microgrids, and power purchase agreements [6]. - The "green electricity direct connection" model allows renewable energy sources to supply electricity directly to users, bypassing the public grid [6]. - The National Development and Reform Commission has issued notifications to promote the development of green electricity direct connection projects [6]. Group 5: Regional Development Initiatives - Various provinces in China are launching plans for green electricity parks, with Shanxi planning to establish 13 pilot parks across multiple cities [7]. - Inner Mongolia aims to identify 20 zero-carbon parks for development, targeting the creation of national-level zero-carbon park pilots [7]. Group 6: Site Selection and Economic Viability - The economic viability of green electricity direct connection projects decreases significantly as the distance between computing centers and power plants increases beyond 100-150 kilometers [8]. - The site selection for computing power centers must balance proximity to energy sources and the demand for low-latency connections [9]. - The Dazhong Park in Datong was chosen for its optimal wind-solar complementarity, robust grid conditions, and government support [9]. Group 7: Cost Reduction Strategies - The initial investment for green electricity direct connection projects is high due to the need for supporting infrastructure [11]. - The Dazhong Park project includes significant investments in solar and wind energy, with a total investment of approximately 3.2 billion yuan [11]. - User-side energy storage in computing centers can enhance electricity stability and reduce costs through various strategies [12]. Group 8: Future Outlook - The demand for green electricity in computing centers is expected to expand further, with a projected increase in green electricity consumption in the computing rental market [6][12]. - The construction of extensive transmission networks will facilitate the synergy between computing and electricity, ensuring stable and green power supply [13].

Core Insights - The rapid development of the new energy industry in China has highlighted the issue of energy consumption, with Jilin Province's Baicheng City strategically expanding its green energy applications, particularly in green hydrogen and new energy storage [1][2]. New Energy Development - Baicheng City has a planned new energy development scale exceeding 16 million kilowatts by the end of 2024, ranking first in Northeast China and among the top ten nationwide [1]. - The city is exploring an integrated model of "green electricity - green hydrogen - green chemicals" to enhance local energy consumption and strengthen the power grid's export capacity [1][2]. Green Hydrogen Projects - Two green hydrogen chemical projects launched in July 2023 have garnered significant attention, achieving multiple national and global milestones [2]. - The projects utilize green electricity from wind and solar sources to produce green hydrogen and subsequently synthesize it into green ammonia, employing a full industrial chain design [2][3]. Technological Innovations - The projects include the world's largest single production facility for green ammonia at 180,000 tons per year and the largest alkaline and PEM mixed electrolysis water hydrogen production system [3]. - They also feature the largest scale of direct current microgrid hydrogen production and solid-state hydrogen storage, addressing challenges in hydrogen storage and production stability [3]. Green Methanol Production - A new integrated project for producing green methanol from wind power and biomass gasification has been launched, with an annual production capacity of 50,000 tons [4]. - This project aims to create a production system of "green electricity - green hydrogen - green methanol," promoting local energy consumption and utilizing agricultural waste [4]. Energy Storage Initiatives - Baicheng City is actively developing energy storage systems, including lithium iron phosphate battery systems and all-vanadium flow batteries, to enhance energy consumption efficiency [7][9]. - The city has initiated several energy storage projects, including a 350 MW compressed air energy storage project, marking a significant step in energy storage technology [10]. Grid Infrastructure Enhancements - The construction of a 100 MW tower molten salt solar thermal power station is underway, which will improve the region's energy export capabilities [12][13]. - Recent upgrades to the power grid, including the completion of multiple 500 kV substations, are designed to support the integration and export of renewable energy projects [14][15]. Future Prospects - The "Jilin Electricity to Beijing" project is in the planning stages, aiming to build an ±800 kV high-voltage direct current transmission line to export 6 million kilowatts of renewable energy from Baicheng to North China [15].

Group 1 - Qinghai Turing Town officially launched in Xining as a key green computing project in Qinghai Province, representing the province's commitment to developing the green computing industry [1] - The project is developed by H3C Group and aims to create an AI innovation community that integrates government, enterprises, innovation, and applications [1] - Qinghai's abundant water, solar, and wind resources provide a strong foundation for the development of clean energy and computing industries, with the province ranking fifth in water resources and second in solar radiation in China [1] Group 2 - Qinghai has implemented policies to transform its resource advantages into a clean energy development framework, with a focus on hydropower and large-scale clean energy bases [3] - As of September, Qinghai's clean energy installed capacity reached 72.52 million kilowatts, accounting for 93.8% of the province's total power capacity, with both installed capacity and generation exceeding 90% [3] - The computing industry in Qinghai is characterized as an eco-friendly industry with high energy capacity and near-zero emissions, which can help address green electricity consumption challenges and support AI and advanced manufacturing sectors [3] Group 3 - China Unicom's Sanjiangyuan National Big Data Base has invested 2.6 billion yuan, with a current computing capacity of 9,000 P, expected to reach 13,000 P by the end of the year [5] - The data center operates with high energy efficiency, achieving a Power Usage Effectiveness (PUE) below 1.2, which is 40% lower than the national average [5] - The Sanjiangyuan green electricity intelligent computing microgrid project integrates renewable energy sources with computing infrastructure, achieving 100% green electricity consumption and zero carbon emissions [5] Group 4 - As of September, Qinghai has built approximately 190,000 standard racks, with computing capacity increasing nearly 40 times in one year [5] - The province's green computing investment constitutes 40% of the fixed asset investment in the industrial park, with over 80% of computing projects powered by green electricity [5] - The establishment of a clean energy and green computing dispatch center has enabled real-time monitoring of energy generation and carbon emissions across the province [5]

新华财经北京11月6日电（记者安娜） 当前全球能源转型与数字革命"双浪叠加"，算力与电力协同发展 备受关注。未来算电协同将如何演进？相关产业发展前景如何？ 在近日于北京举办的2025第三届电力数智化大会上发布的《电力与算力协同发展蓝皮书（2025）》给出 答案。 蓝皮书指出，绿色低碳是当今全球能源转型的核心主线，而算力已成为数字经济的关键生产力。电力与 算力深度协同，既是实现碳达峰碳中和目标的重要路径，也是推动新型能源体系构建的核心支点。 但国内算电协同仍面临多重挑战。记者从会上了解到，当前，我国算力中心与清洁能源基地尚未形成紧 密联动；算力优化与电力调度缺乏统一的算法与协议框架；算电一体化的指标体系、度量体系和交易机 制尚未全面建立；算力、电力、碳市场之间的价格信号传导与利益分配机制也有待健全。这些问题的存 在，使算电协同在效率、绿色与安全三方面均待进一步突破。 尽管如此，蓝皮书认为，算电协同在国内发展前景依然广阔。一方面，自2023年以来，《算力基础设施 高质量发展行动计划》《加快构建新型电力系统行动方案（2024-2027年）》及《关于深入实施"人工智 能+"行动的意见》等政策相继出台，为算电一体化提供 ...

Core Insights - China's computing power and electricity have been rapidly developing, with total computing power expected to exceed 280 EFlops by the end of 2024, averaging a growth rate of nearly 30% over the past five years [1] - Annual electricity consumption of data centers has surpassed 150 billion kilowatt-hours, highlighting the increasing demand for energy [1] - The structural contradiction between the over 50% share of renewable energy installations in the western region and the electricity shortages in the eastern region is becoming more pronounced [1] - The dual pressure of surging computing demand and energy transition necessitates a deep integration of computing power and electricity [1] - The essence of "computing-electricity synergy" is to break down the barriers between bits (computing power) and watts (electricity), allowing computing power to act as a "regulator" in the electricity system [1] - This synergy aims to promote renewable energy consumption, reduce electricity costs, and optimize resource allocation through the two-way matching of computing load and electricity supply [1]

Core Insights - The article emphasizes the growing importance of computing power as a core production factor, comparable to water and electricity, with energy being the foundation for its continuous operation [1] - The collaboration between DaMao Technology and CATL aims to address the challenges of energy supply and demand imbalance through AI-driven solutions [3][6] Group 1: Industry Trends - By 2024, China's data center electricity consumption is projected to reach 166 billion kWh, accounting for 1.68% of total electricity usage, with power consumption per unit computing power at approximately 5.9 kWh/EFLOPS [1] - The increasing demand for computing power is leading to mismatches in energy supply and demand, highlighting the need for intelligent energy management solutions [1][3] Group 2: Company Developments - DaMao Technology has secured nearly 100 million yuan in A+ round financing to develop core technologies for energy modeling and collaborative computing power management [1][6] - The company has successfully implemented its energy management solutions in key projects, achieving a 6.5% reduction in electricity costs and enhancing the self-balancing capability of the power system [6] Group 3: Technological Innovations - DaMao Technology's approach involves creating a digital twin relationship between computing tasks and energy supply, utilizing AI, IoT, and blockchain technologies for optimal energy scheduling [3][4] - The company has developed a decision-making system that predicts energy demand based on historical data, allowing for precise energy dispatching [4] Group 4: Market Positioning - DaMao Technology aims to capture a 30% market share in the collaborative computing power sector within three years, establishing itself as a leader in the industry [7] - The company has formed strategic partnerships with major players like Schneider Electric and CATL, enhancing its market presence and technological capabilities [7][8]

Core Viewpoint - The article emphasizes the growing importance of computing power as a core production factor, comparable to water and electricity, with energy being the foundation for its continuous operation [1] Group 1: Energy Consumption and Demand - By 2024, China's data center electricity consumption is projected to reach 166 billion kWh, accounting for 1.68% of the total electricity consumption in society, with power consumption per unit of computing power at approximately 5.9 kWh/EFLOPS [1] - The increasing demand for computing power is leading to mismatches in energy supply and demand, highlighting the need for intelligent solutions for precise energy scheduling and efficient utilization [1][3] Group 2: AI-Driven Energy Coordination - The concept of "computing power and electricity coordination" is introduced, focusing on balancing computing demand with energy supply through AI-driven models [3][4] - The company aims to match the "computing power consumption curve" with the "power generation supply curve" to achieve this balance [4] Group 3: Company Overview and Achievements - Founded in 2021, the company has become a star enterprise in Shanghai, focusing on energy management in computing centers, integrating AI technology for dynamic decision-making [5] - The company's technology has been validated in benchmark projects, achieving a 6.5% reduction in electricity costs and enhancing the self-balancing capability of the power system [6] Group 4: Market Position and Future Plans - The company plans to capture a 30% market share in the domestic computing power and electricity coordination business within three years, supported by strategic partnerships with major energy management firms [9] - The company is also exploring international markets, including Japan and Southeast Asia, to expand its consulting and lifecycle service projects [9]

Core Insights - Shanghai Damou Technology Co., Ltd. has completed nearly 100 million yuan in A+ round financing, with the investment coming from Puquan Capital, the only industrial investment platform under CATL [1] - The company, established in 2021, is the first in China to apply energy large models to collaborative electricity calculation, virtual power plants, and pan-energy scenarios, indicating CATL's strategic shift from an "energy equipment manufacturer" to a "zero-carbon energy manager" [1][2] - The energy industry is accelerating its transition towards intelligence and digitalization, with AI technology becoming a key tool to address the challenges posed by the increasing share of renewable energy [2] Group 1: Industry Trends - China's power system is transitioning from a traditional "source-following-load" model to an "interactive source-grid-load-storage" model due to the rapid increase in renewable energy [2] - The National Development and Reform Commission and the Energy Administration have set goals for integrating AI with energy systems, aiming to establish a fusion innovation system by 2027 and achieve global leadership in key technologies by 2030 [2] Group 2: Company Developments - Damou Technology has developed a self-adaptive energy operation system to reduce operational costs of computing centers and promote clean energy consumption, achieving over 90% prediction accuracy and 95% decision accuracy [3][5] - The company has successfully implemented its technology in several demonstration projects, including a project with Fujian Big Data Group that improved energy operation revenue by over 25% [5] Group 3: Strategic Value of Financing - The recent financing will be used for the independent research and development of core technologies related to energy large models and collaborative electricity calculation platforms, solidifying Damou Technology's leading position in this field [7] - The investment from CATL represents a strategic move towards becoming a comprehensive energy solution provider, integrating real-time sensing, prediction, decision-making, and optimization capabilities [7][8] - Prior to CATL's investment, Damou Technology had already secured strategic investments from SenseTime and Cambrian, creating a complete collaborative ecosystem across the industry chain [8]