转自：中国科学报 近年来，人工智能（AI）、机器学习、云计算及大数据技术呈现出爆炸式发展态势，全球数据处理需 求呈指数级增长。数据中心作为数字经济的核心物理载体，其能耗问题日益严峻。 1月5日，《自然综述-电气工程》在线发表了华南理工大学电力学院教授朱继忠团队与合作者的最新研 究成果。他们提出了一种以海洋可再生能源为支撑的水下数据中心零碳运行架构，为高能耗AI基础设 施的绿色转型提供了全新解决方案。 "这项工作是当前'绿色AI'背景下的一次前瞻性探索。"论文通讯作者朱继忠对《中国科学报》表示，该 研究首次系统性设计并论证了融合风能、光伏、波浪能和海水冷却的水下数据中心能源与热管理系统框 架，验证了其实现"全天候零碳运行"的可行性，展现出推动AI计算基础设施绿色化、低碳化的潜力。 《自然综述-电气工程》的审稿人对这一研究成果给予了高度评价。他们认为，该研究"在绿色计算基础 设施设计方面提供了系统性新思路"，兼具"突破性技术创新和工程可行性"。 数据中心绿色转型迫在眉睫 国际能源署（IEA）2025年4月发布的报告显示，2024年全球数据中心的用电量约为415太瓦时（TWh， 1太瓦时=10亿千瓦时），占全球电力 ...

为推动技术协同与合作落地，大会同期举办"AI大模型·新智驱"专题座谈会，新紫光集团、科大讯飞等 近20家生态企业代表与北汽研究总院、北汽产投核心团队，围绕AI大模型在汽车产业的前瞻应用、产 业链协同需求等关键议题展开深度研讨。现场举行的多项签约与发布仪式，成为北汽科创生态协同发展 的生动实践。 当前，汽车产业正处于新一轮科技革命与产业变革的交汇点，智能化、新能源成为核心发展方向。作为 我国汽车工业的骨干力量，北汽集团积极发挥产业引领作用。 近日，北汽集团科创投资生态论坛暨北汽产投生态伙伴大会在北京举行。来自产业界、投资界、科技领 域的200余位生态伙伴齐聚一堂，以"聚变而升"为主题，围绕新质生产力培育、汽车产业智能化绿色化 转型等核心议题，搭建起技术交流、资本对接、合作共赢的高端平台，为中国汽车产业高质量发展凝聚 共识、汇聚力量。 据北汽方面介绍，本次论坛是北汽产投深化生态布局、赋能产业升级的关键举措，既集中展现了北汽与 生态伙伴在核心技术领域的前沿成果，更释放出以资本为纽带、以技术为核心、以生态为载体的开放发 展信号。 北汽集团副总经理刘宇为大会作开场致辞，北汽产投作为北汽集团产业投资与资本运作平台，肩负在 ...

2025年12月16日，北汽集团科创投资生态论坛暨北汽产投生态伙伴大会在北京举行。来自产业界、投资界、科技领域的200余位生态伙伴齐聚一堂， 以"聚变而升"为主题，围绕新质生产力培育、汽车产业智能化绿色化转型等核心议题，搭建起技术交流、资本对接、合作共赢的高端平台，为中国汽车产业 高质量发展凝聚共识、汇聚力量。 大、范围更全的产业级供应体系，实现多方共赢。 为推动技术协同与合作落地，大会同期举办"AI大模型・新智驱"专题座谈会，新紫光集团、科大讯飞等近20家生态企业代表与北汽研究总院、北汽产投 核心团队，围绕AI大模型在汽车产业的前瞻应用、产业链协同需求等关键议题展开深度研讨。现场举行的多项签约与发布仪式，成为北汽科创生态协同发 展的生动实践。 大会邀请多位权威嘉宾带来深度分享，以多元视角解码产业发展逻辑。世界金融论坛高级研究员、中国银行原首席研究员宗良从全球经济格局切入，解 读汽车产业发展趋势；中国银河证券首席经济学家章俊聚焦大国规模优势，剖析绿色AI的发展逻辑与产业机遇；星动纪元创始人、清华大学交叉信息研究 院助理教授陈建宇分享AI具身智能最新进展，揭示其赋能实体产业的关键路径。 北汽集团相关业务板块负责 ...

Core Viewpoint - The 20th Central Committee of the Communist Party of China emphasizes accelerating high-level technological self-reliance and strength, with technology finance serving as a crucial support for technological and industrial innovation, driving the development of new productive forces [1] Group 1: Pain Points and Challenges in Technology Finance - The development of technology finance has faced structural obstacles, including information asymmetry, insufficient linkage between debt and equity financing, and the need for improved efficiency in service delivery throughout the lifecycle of technology enterprises [2][3][4][5][6] Group 2: AI Empowerment in Technology Finance - AI technology offers a new path to address existing challenges by enhancing data processing and pattern recognition capabilities, enabling dynamic evaluation of enterprises' true operational status and core technological strength [1][7] - AI can create precise enterprise profiles and optimize investment research decisions, facilitating the discovery and dynamic assessment of "technological credit" [1][7][10] Group 3: Key Paths for AI Empowerment - The core path of AI empowerment in technology finance involves using AI to drive precise profiling and credit reconstruction of enterprises, enabling efficient matching of financial resources based on dynamic risk assessments [7][14] - AI enhances the identification and prediction of risks associated with "technological credit," integrating risk assessment into the financial system [11][12] Group 4: Enhancements in Financing Mechanisms - AI facilitates adaptive matching of financial resources for both debt and equity financing, allowing for tailored financial solutions based on the lifecycle and risk characteristics of technology enterprises [14][15][16] - The integration of AI in investment processes improves the efficiency of due diligence and enhances the accuracy of investment decisions [15][16] Group 5: Capital Market Enhancements - AI transforms non-standard and illiquid "technological credit" into standardized and highly liquid financial assets, enhancing the operational efficiency and quality of capital markets [17][18] - AI can improve market services and inclusivity by providing deep analysis and valuation references for under-researched companies, thus attracting long-term capital [18][19] Group 6: Recommendations for Future Development - The industry should focus on strengthening green AI applications, enhancing data infrastructure, cultivating interdisciplinary talent, and establishing comprehensive risk governance paths to support the sustainable development of technology finance [20][21][22][23][24]

Core Insights - The forum highlighted the urgent energy challenges posed by AI, particularly in relation to data centers' increasing electricity and water consumption [9][10] - A significant focus was placed on the need for sustainable practices in AI and data center operations, emphasizing the importance of balancing innovation with sustainability [10][11] Group 1: Energy and Water Consumption - By 2024, data centers are projected to account for approximately 1.5% of global electricity consumption, reaching around 415 terawatt-hours (TWh) [8] - By 2030, global data center electricity demand is expected to more than double to about 945 TWh, and by 2035, it could rise to approximately 1200 TWh [8] - Google's data centers consumed 24.2 billion liters of water in 2023, equivalent to the water volume of 1.7 West Lakes [8] Group 2: Liquid Cooling Technology - Liquid cooling technology is gaining traction due to its ability to enhance cooling efficiency by 1000-3000 times compared to traditional methods [14] - The Chinese liquid cooling server market is projected to reach $2.37 billion in 2024, with a year-on-year growth of 67% [14] - The market is expected to grow at a compound annual growth rate (CAGR) of 46.8% from 2024 to 2029, reaching $16.2 billion by 2029 [14] Group 3: Lifecycle Water Management - Ecolab's Nalco brand has developed a comprehensive lifecycle water management solution for liquid cooling data centers, addressing challenges such as cooling liquid quality and maintenance [15][22] - The implementation of lifecycle management practices has significantly improved system efficiency and reduced operational risks, as demonstrated by a case study where water change frequency increased from bi-weekly to over a year [22][23] - The overall value benefits from improved management practices in liquid cooling data centers can exceed millions, enhancing both operational efficiency and sustainability [22][23]