Multiobjective optimization of dynamic aperture
Multiobjective optimization of dynamic aperture
复制标题
DOI：
10.1103/physrevstab.14.054001
复制
发表时间：
2011-05
影响因子：
--
通讯作者：
Lingyun Yang;Y. Li;W. Guo;S. Krinsky
中科院分区：
文献类型：
--
作者：
Lingyun Yang;Y. Li;W. Guo;S. Krinsky
动态孔径是储存环的重要非线性特性之一。虽然已经有解析和数值方法来找到孔径，如何优化它的反向问题仍然是一个具有挑战性的问题。在加速器的设计和运行中，需要一种通用的、灵活的优化DA的方法。在本文中，我们讨论了使用多目标优化DA。首先，我们考虑使用目标函数的基础上，只有数值跟踪结果。这些结果的数据挖掘表明DA和低阶非线性驱动项之间的相关性。接下来，我们考虑使用目标函数，其中包括数值跟踪结果和低阶非线性驱动项的分析估计。这导致了更快的收敛。以国家同步辐射光源II（NSLS-II）晶格为例，说明了该方法的应用。这种多目标方法不受特定的线性或非线性晶格设置的限制，也可以应用于优化存储环的其他属性。
Dynamic aperture (DA) is one of the key nonlinear properties for a storage ring. Although there have been both analytical and numerical methods to find the aperture, the reverse problem of how to optimize it is still a challenging problem. A general and flexible way of optimizing the DA is highly demanded in accelerator design and operation. In this paper, we discuss the use of multiobjective optimization for DA. First we consider using objective functions based only on numerical tracking results. Data mining of these results demonstrated a correlation between DA and low-order nonlinear driving terms. Next we considered using objective functions which included both numerical tracking results and analytical estimates of low-order nonlinear driving terms. This resulted in faster convergence. The National Synchrotron Light Source II (NSLS-II) lattice was taken as an example to illustrate this method. This multiobjective approach is not limited by particular linear or nonlinear lattice settings, and can also be applied for optimizing other properties of a storage ring.