Speaker
Description
A fully automated beam dynamics design method based on genetic algorithm is proposed for a 17 GHz X-band three-cell standing wave accelerating structure to meet the miniaturization requirements of X-ray based intraoperative radiotherapy(IORT) devices. The genetic algorithm simultaneously optimizes the field distribution of the three cavities, achieving both electron beam bunching and acceleration within a single framework without manual intervention. The design results demonstrate that the 8~keV electron bunch is accelerated to a target energy of 50~keV, satisfying the constraints of IORT equipment. The six-dimensional design space, comprising the peak field ((E_1, E_2, E_3)) and the axial length ((L_1, L_2, L_3)) of the three cells, is searched with a genetic algorithm (GA) coupled to the RF-Track particle-in-cell tracking engine. The fitness function integrates multiple parameters including the target energy to ensure the optimization results meet the requirements. After 100 generations, the GA reaches a best fitness of 123.73, with 76.57$\%$ of the particles inside the 48-51keV acceptance band, and a transmission of 93.89$\%$.
| I have read and accept the Privacy Policy Statement | Yes |
|---|