Model-independent particle accelerator tuning
Model-independent particle accelerator tuning
复制标题
DOI：
10.1103/physrevstab.16.102803
复制
发表时间：
2013-10
影响因子：
--
通讯作者：
A. Scheinker;X. Pang;L. Rybarcyk
中科院分区：
文献类型：
--
作者：
A. Scheinker;X. Pang;L. Rybarcyk
洛斯阿拉莫斯国家实验室，美国新墨西哥州洛斯阿拉莫斯87545(2013年7月25日收到；2013年10月21日发表)我们提出了一种新的独立于模型的动态反馈技术，转速调节，用于自动并同时调节不确定、复杂系统的耦合部件。该方法的主要优点是：(1)它具有处理未知、时变系统的能力；(2)它给出了参数更新率的已知界限；(3)我们给出了它的收敛和稳定性的解析证明；(4)它通过实验物理与工业控制系统(EPICS)这样的控制系统具有简单的数字实现。由于该技术与模型无关，因此对于不确定和时变系统，它可能是一种实时的、硬件内的、基于反馈的优化方案。特别是，它足够坚固，可以处理由于耦合、热循环、不对准和制造缺陷造成的不确定性。因此，它可作为现有加速器调谐/控制方案的fi新调谐补充。我们给出的多粒子模拟结果表明，当束流性质和射频相移连续变化时，该方案能够同时自适应地调整洛斯阿拉莫斯中子科学中心(LANSCE)直线加速器传输区中22个四极磁体和双聚束器腔的设定值。调谐仅基于束流读数，而不了解粒子动力学。我们还概述了如何在软件中实现这一通用方案以进行优化，并在硬件中实现基于反馈的控制/调谐，以用于广泛的系统。
Los Alamos National Laboratory, Los Alamos, New Mexico 87545, USA(Received 25 July 2013; published 21 October 2013)We present a new model-independent dynamic feedback technique, rotation rate tuning, for automati-cally and simultaneously tuning coupled components of uncertain, complex systems. The main advan-tages of the method are: (1) it has the ability to handle unknown, time-varying systems, (2) it gives knownbounds on parameter update rates, (3) we give an analytic proof of its convergence and its stability, and(4) it has a simple digital implementation through a control system such as the experimental physics andindustrial control system (EPICS). Because this technique is model independent it may be useful as a real-time, in-hardware, feedback-based optimization scheme for uncertain and time-varying systems. Inparticular, it is robust enough to handle uncertainty due to coupling, thermal cycling, misalignments,and manufacturing imperfections. As a result, it may be used as a fine-tuning supplement for existingaccelerator tuning/control schemes. We present multiparticle simulation results demonstrating thescheme’s ability to simultaneously adaptively adjust the set points of 22 quadrupole magnets and tworf buncher cavities in the Los Alamos Neutron Science Center (LANSCE) Linear Accelerator’s transportregion, while the beam properties and rf phase shift are continuously varying. The tuning is based only onbeam current readings, without knowledge of particle dynamics. We also present an outline of how toimplement this general scheme in software for optimization, and in hardware for feedback-based control/tuning, for a wide range of systems.