Researchers have developed a surrogate model named FIREWALL (Fast Integrated Runaway Electron WALL loads) to accelerate the assessment of wall damage from runaway electron (RE) beams in tokamaks. During plasma disruptions, REs can create intense, localized heat loads on plasma-facing components, risking structural melting. The new model, detailed in a preprint, provides a computationally efficient alternative to full-scale Monte Carlo simulations, which are often too slow for comprehensive scenario scanning or device-scale assessments. FIREWALL is designed to quickly translate global RE simulation data into predictions of wall melting across complex three-dimensional geometries, identifying high-risk areas for further analysis. Source: arXiv
The primary challenge in mitigating RE damage is the immense computational cost of existing high-fidelity models. These workflows typically couple particle transport codes like extsc{Geant4} with three-dimensional thermomechanical response modeling to quantify material damage. While accurate, their demand on computing resources makes it impractical to evaluate the wide range of potential disruption scenarios a device like ITER might face. This computational bottleneck limits the ability to proactively design mitigation strategies and optimize operational parameters to avoid the most dangerous RE-wall interactions. FIREWALL directly addresses this gap by significantly reducing the time needed for a global assessment. Source: arXiv
The primary challenge in mitigating RE damage is the immense computational cost of existing high-fidelity models.
FIREWALL's architecture combines a pre-calculated database of volumetric energy-deposition profiles from extsc{Geant4} with a one-dimensional heat-diffusion solver applied to each wall element. This approach retains critical physics, including the energy and incident angle distributions of the impacting runaway electrons, which are essential for accurate heat load calculations. By simplifying the thermal response to a 1D problem for each discrete surface area while using a database of 3D interaction results, the model can rapidly compute the spatiotemporal temperature evolution of the entire wall. This method allows it to predict temperatures up to the material's melting threshold without the overhead of a full 3D thermomechanical simulation for every scenario. Source: arXiv
The model's primary function is to serve as a screening tool. By enabling large-scale parametric scans of disruption scenarios, operators and designers can identify the specific conditions and wall locations most vulnerable to RE-induced melting. This allows for the strategic deployment of more resource-intensive, high-fidelity simulations, focusing them only on the limited regions identified as critical by FIREWALL. This two-tiered approach—rapid, broad assessment followed by targeted, detailed analysis—optimizes the use of computational resources and provides a more comprehensive framework for ensuring the integrity of tokamak plasma-facing components in future fusion reactors. Source: arXiv