Imagine trying to predict how a candle melts, but the candle is inside a high-pressure reactor. Instead of doing expensive and dangerous real-world tests, this project creates a super-accurate 3D digital map of nuclear fuel. It uses powerful computers to simulate exactly how materials wear down and react to heat, making reactors safer and more reliable.
Licensing new nuclear fuels is slow and expensive because current 1D simulation tools cannot accurately predict 3D thermo-mechanical behavior. This creates a bottleneck for deploying accident-tolerant fuels and extending reactor lifespans.
A suite of open-source 3D high-fidelity simulation tools and AI-based reduced order models for nuclear fuel elements.
If you are a plant operator dealing with the need to extend the life of current reactors while changing the electricity mix — this project developed 3D simulation tools that provide a detailed description of fuel behavior. This allows for safer operation and better safety assessments under new conditions.
If you are a fuel designer dealing with the slow and costly licensing of new materials — this project developed open-source fuel performance codes. These tools speed up the verification and validation process required by safety authorities for innovative fuel designs.
If you are a software provider dealing with the limitations of 1D models — this project developed AI-driven reduced order models and 3D reference databases. This allows for the integration of high-fidelity physics into faster, industrial-grade simulation software.
Based on available project data, the tools are being developed as open-source fuel performance codes, suggesting a non-commercial distribution model for the core software.
Yes, the project specifically focuses on transferring knowledge from high-fidelity 3D simulations to industrial fuel performance codes using AI and reduced order models.
The project explicitly mentions the development of open source fuel performance codes, though specific licensing terms for the resulting AI models are not detailed in the summary.
Yes, it aims to meet safety authority requirements regarding the Verification, Validation, and Uncertainties Quantification process for licensing innovative fuels.
The project period runs from 2022-11-01 to 2027-04-30, meaning full results are expected by April 2027.
The consortium is well-balanced for technology transfer, featuring 19 partners across 11 countries. With a 37% industry ratio (7 industrial partners, including 2 SMEs), there is a strong bridge between the 12 academic/research entities and the commercial market, ensuring the 3D simulation tools are grounded in industrial needs.
Contact the Commissariat à l'énergie atomique et aux énergies alternatives (CEA) in France.
Contact us to identify the specific AI-driven reduced order models available for your fuel simulation pipeline.
