en.Wedoany.com Reported - An Austrian research team has proposed a new concept, the "AI-energy storage nexus," arguing that the rapid growth of AI data centers combined with large-scale energy storage deployment can accelerate the clean energy transition. The related paper was recently published in the journal Energy and Climate Change.
"Contrary to common expectations, AI growth is creating opportunities to put the clean energy transition on track," said researcher Behnam Zakeri, attributing this primarily to the nexus effect between AI and energy storage. The two technologies are forming a virtuous cycle in which the development and scaling of one technology positively impacts the other. The research team identified where these synergies occur and coined the term "AI-energy storage nexus" for the first time.
Zakeri stated that energy storage can address multiple issues in the growth of digital infrastructure, not only in terms of renewable energy matching, but also in shortening grid connection times and providing flexibility for the phased grid integration of data centers. Investment from major tech companies is shifting toward energy storage solutions, which will advance the R&D and deployment of related technologies and generate spillover effects for the broader energy transition beyond digital infrastructure.
The research shows that machine learning and generative AI can accelerate the discovery of molten salt materials for thermal energy storage, identify new battery chemistries, and improve battery lifespan and recyclability. Meanwhile, the growing demand from AI data centers is helping long-duration energy storage move toward commercialization, with Google having signed offtake agreements for liquid CO2 and iron-air energy storage.
The pathways through which energy storage supports AI infrastructure include: reducing peak grid imports; alleviating grid connection constraints and deferring costly transmission upgrades; improving power quality and reliability through backup power; enabling demand response, allowing data centers to reduce demand or export power during grid emergencies and increasing self-consumption of on-site renewables; and enhancing operational flexibility through workload scheduling, load balancing, and other forms of flexible computing.
In their conclusions, the paper's authors put forward four policy recommendations: reform grid connection rules to reward data centers that are flexible and grid-supportive; remove barriers to co-locating battery storage with renewable energy; incentivize grid-aware computing and demand-side flexibility; and accelerate long-duration energy storage deployment through targeted procurement and financing policies.
Zakeri added that the team currently plans to develop operational models for energy storage co-located with data centers or located near the grid, in order to address some of the challenges facing today's power grids. The paper was completed through collaboration among researchers from the Vienna University of Economics and Business (WU), the International Institute for Applied Systems Analysis (IIASA), and King Abdullah University of Science and Technology in Saudi Arabia.
