Speaker
Description
We propose an AI-driven control architecture that maintains resonance between magnetrons and SRF cavities, enabling efficient accelerator operation. Injection locking is achieved by extracting pick-up signals from normal-conducting cavities integrated into a beamline with the main SRF cavities; a bunched electron beam excites these resonators, providing decoupled RF power to phase-lock the magnetrons. This approach eliminates the need for external RF sources for phase locking. AI algorithms optimize cavity tuning to preserve the phase-locked condition, while AI-controlled ferroelectric tuners (FRTs) compensate for microphonics and other detuning effects. We emphasize the complexity of FRT control, which requires both fast (sub-millisecond) bias-voltage signals and slower temperature-driven adjustments via an integrated chiller. Fast control addresses microphonics compensation, whereas temperature control extends the tuning range to mitigate slow drifts. We therefore propose the development and deployment of high-fidelity, AI-driven digital twins to enable dynamic, in-situ linac control, reducing system cost, improving beam quality, and increasing reliability.
| I have read and accept the Privacy Policy Statement | Yes |
|---|