Autonomous Pressure Control in MuVacAS via Deep Reinforcement Learning and Deep Learning Surrogate Models

TL;DR

This paper introduces a data-driven, autonomous pressure control system for MuVacAS using a deep learning surrogate model and deep reinforcement learning, enabling precise regulation in a complex accelerator environment.

Contribution

It presents a novel combination of a deep learning surrogate model and reinforcement learning for autonomous pressure control in accelerator systems.

Findings

The surrogate model accurately emulates the argon injection dynamics.

The reinforcement learning agent effectively maintains pressure within operational limits.

The approach enables autonomous control in complex, dynamic environments.

Abstract

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.