One of the primary challenges for future nuclear fusion power plants is understanding how neutron irradiation affects reactor materials. To tackle this issue, the IFMIF-DONES project aims to build a facility capable of generating a neutron source in order to irradiate different material samples. This will be achieved by colliding a deuteron beam with a lithium jet. In this work, within the DONES-FLUX project, deep learning surrogate models are applied to the design and optimization of the IFMIF-DONES linear accelerator. Specifically, neural operators are employed to predict deuteron beam envelopes along the longitudinal axis of the accelerator and neutron irradiation effects at the end, after the beam collision. This approach has resulted in models that are able of approximating complex simulations with high accuracy (less than 17% percentage error for the worst case) and significantly reduced inference time (ranging from 2 to 6 orders of magnitude) while being differentiable. The substantial speed-up factors enable the application of online reinforcement learning algorithms, and the differentiable nature of the models allows for seamless integration with differentiable programming techniques, facilitating the solving of inverse problems to find the optimal parameters for a given objective. Overall, these results demonstrate the synergy between deep learning models and differentiable programming, offering a promising collaboration among physicists and computer scientists to further improve the design and optimization of IFMIF-DONES and other accelerator facilities. This research will lay the foundations for future projects, where optimization efforts with differentiable programming will be performed.
Developing materials capable of withstanding extreme irradiation is a fundamental challenge for realizing nuclear fusion as a sustainable energy source. The IFMIF–DONES facility, currently under construction, will address this need by producing a fusion-relevant neutron flux using a high-power deuteron beam striking a flowing lithium target, enabling accelerated qualification of candidate structural materials. Validating the accelerator and target technologies requires precise control of the ultra-high-vacuum region near the lithium surface, yet direct pressure sensing in this irradiated zone is challenging and conventional controllers can struggle with the associated nonlinear gas-flow dynamics. As part of the DONES–FLUX project, this work introduces a data-driven workflow for indirect pressure estimation and autonomous regulation in the MuVacAS prototype, a dedicated experimental platform that reproduces the final beamline section of IFMIF–DONES. Two Fourier Neural Operator surrogate models are learned from real data: one serves as a virtual sensor to infer chamber pressure from upstream readings, while the other models system dynamics to provide a safe, fast training environment for a Deep Reinforcement Learning controller. The trained agent uses virtual-sensor feedback during deployment and reliably regulates argon injection across operating scenarios, including conditions not seen during training. Experimental campaigns show that the RL-based controller yields promising control results, achieving stable regulation without manual retuning. These results demonstrate a viable pathway for deploying advanced learning-based control in fusion-relevant accelerator systems, supporting the operational readiness of IFMIF–DONES and future fusion technology infrastructures.
The development of nuclear fusion requires materials that can withstand extreme conditions. The IFMIF-DONES facility, a high-power particle accelerator, is being designed to qualify these materials. A critical testbed for its development is the MuVacAS prototype, which replicates the final segment of the accelerator beamline. Precise regulation of argon gas pressure within its ultra-high vacuum chamber is vital for this task. This work presents a fully data-driven approach for autonomous pressure control. A Deep Learning Surrogate Model, trained on real operational data, emulates the dynamics of the argon injection system. This high-fidelity digital twin then serves as a fast-simulation environment to train a Deep Reinforcement Learning agent. The results demonstrate that the agent successfully learns a control policy that maintains gas pressure within strict operational limits despite dynamic disturbances. This approach marks a significant step toward the intelligent, autonomous control systems required for the demanding next-generation particle accelerator facilities.