电
模块化设计
网格
环境经济学
风力发电
可再生能源
化石燃料
环境科学
发电
碳中和
消费(社会学)
高效能源利用
电池(电)
工程类
计算机科学
能源消耗
文件夹
电力市场
能源
能量(信号处理)
能量建模
情景分析
独立电源系统
电力系统
市电
温室气体
微电网
光伏
碳足迹
能源技术
作者
Apoorv Lal,Siyuan Wang,Fengqi You
标识
DOI:10.1021/acs.est.6c09330
摘要
The rapid expansion of AI infrastructure in the US is emerging as a major new source of electricity demand, raising concerns about fossil lock-in during accelerated grid expansion. Here, we assess the role of small modular reactors (SMRs) within portfolios that also include on-site solar and wind power, grid electricity, battery storage, and direct air capture. Using a state-resolved scenario analysis for 2028–2035, we compare nuclear-inclusive and non-nuclear portfolios across alternative AI load trajectories, SSP-RCP electricity mixes, and decarbonization constraints. Relative to a grid-only counterfactual, nuclear-inclusive portfolios that achieve carbon neutrality by 2035 require 2.4–47.2 GW of SMR capacity, avoid 0.3–1.3 GtCO2, reduce water consumption by an amount equivalent to the annual use of 28.3–50.5 million people, and deliver \$105.8–194.8 billion in cumulative energy-system cost savings across climate and efficiency pathways. The results show that SMRs can materially expand cost-effective decarbonization options for AI infrastructure, while their contribution depends on timely deployment, permitting, and coordination with state-level electricity planning.
科研通智能强力驱动
Strongly Powered by AbleSci AI