Idaho National Laboratory (INL) and Amazon Web Services (AWS) have established a collaboration to use artificial intelligence to advance nuclear energy technology. The partnership will apply AWS’s cloud computing infrastructure and AI tools to INL’s work on next-generation nuclear reactors.
The primary goal is to develop technologies that reduce the cost and time required to design, license, build, and operate nuclear facilities. The long-term objective is to enable safe and reliable autonomous operation of advanced reactors, accelerating their deployment.
“INL is committed to accelerating nuclear energy deployment, advancing autonomous nuclear operations and leading nuclear energy science through transformational AI technology — achieved with AWS and the nuclear energy industry,” said Chris Ritter, division director of Scientific Computing and AI at INL.
Utilizing AWS’s computing power
Under the agreement, INL will utilize AWS’s advanced computing power, AI foundation models through Amazon Bedrock, and specialized hardware.
“Through this collaboration with AWS, we have access to AI models, GPUs (graphical processing units) and specialized cloud services, including Amazon’s Bedrock service, which will enable INL researchers to use many leading foundation models to build nuclear energy applications,” added Ritter.
INL Director John Wagner stated the partnership is a key step for the nation’s energy goals. “By leveraging AWS’s cutting-edge cloud computing and AI solutions, we can accelerate nuclear energy deployment for America,” he said.
From the industry perspective, AWS views the technology as a way to improve reactor safety and efficiency.
“AWS’s powerful AI and computing technology will support Idaho National Laboratory’s development of autonomous nuclear reactors to pioneer a future where civilian nuclear operations are safer, smarter and more responsive,” said David Appel, vice president for AWS.
Creation of digital twin
A key initial project will be the creation of a digital twin for a small modular reactor (SMR). A digital twin is a virtual model of a physical asset that uses near real-time data to run complex simulations. This is an important step toward using AI for autonomous control systems.
“INL will use AWS Compute and AI tools to develop a digital twin of a small modular reactor — nuclear reactors that range in size from 20 to 300 megawatts of electricity,” INL explained in a press release.
The effort is part of a larger INL strategy to create an ecosystem where government laboratories, AI firms, and nuclear developers can collaborate.
One potential future application for these advanced autonomous reactors is to provide clean, reliable power for large-scale data centers.
Using AI for nuclear energy
There is a growing trend of using AI to boost nuclear energy efforts. In a recent event, The Oak Ridge National Laboratory (ORNL) has penned a memorandum of understanding with AI company Atomic Canyon to refine the licensing process for nuclear power plants.
It aims to use high-performance computing to create simulations that will ensure the safety of designs while accelerating the licensing process with AI.
“With new capabilities enabled by AI and partners like Atomic Canyon, we can help the nuclear industry unleash American energy,” remarked ORNL Director Stephen Streiffer.