Speaker
Description
The beam quality is of utmost importance in the operation of the IRRAD proton irradiation facility at CERN. During irradiation, the intensity, the transverse and longitudinal beam profiles are monitored online with custom-made Beam Profile Monitor (IRRAD-BPM) devices. The BPM sensors were developed and recently significantly improved thanks to a new manufacturing technology based on microfabrication of metal nano-layers. To be able to exploit all features of these new devices, the readout DAQ technology, as well as the handling and display of the BPM data, also needed to be substantially improved. The existing DAQ technology employed for the readout needed to be upgraded to cope with increasing sensitivity (nA range) and timing performance (sampling rate down to the ms range). Moreover, the display and analysis of the BPM data can also be substantially improved with the innovative idea of applying Machine Learning (ML) techniques. A first prototype of an ML model, aiming to perform the automatic pattern recognition or anomaly detection of beam profiles, was thus developed and tested. The first prototype of this full new system is currently being tested during the irradiation run 2026 of IRRAD with the goal to replace the former system after the CERN Long Shutdown 3 (LS3). In this work we present the design and the architecture of the new IRRAD-BPM system (featuring upgraded sensors, DAQ unit and data handling capabilities), and the latest results on its performance with beam.
Funding Agency
Work supported in part by the European Union’s 2020 research and innovation program under grant agreement no. 101057511 (EURO-LABS).
| I have read and accept the Privacy Policy Statement | Yes |
|---|---|
| Paper submission software | MSWord |