强流束晕-混沌的RBF神经网络控制  

Control of High-current Beam Halo-chaos Based on RBF Neural Network

作　　者：高远[1] 袁海英[1] 孔峰[1] 谭光兴[1] 

机构地区：[1]广西工学院电子信息与控制工程系,广西柳州545006

出　　处：《宁夏大学学报（自然科学版）》2011年第3期231-234,共4页Journal of Ningxia University(Natural Science Edition)

基　　金：广西自然科学基金资助项目(2010GXNSFA013126);广西教育厅科研资助项目(201010LX243)

摘　　要：以周期性聚焦磁场通道中的Kapchinskij-Vladimirskij(K-V)分布离子束为对象,研究了强流束晕-混沌现象的RBFNN自适应控制方法.该方法以神经网络的输出作为周期聚焦磁场的线性控制因子,通过对外部磁场的线性调节实现束晕-混沌控制.模拟结果表明:当选择恰当的RBFNN控制结构,自适应调整其内部参数,可将混沌变化的束包络半径控制在匹配半径附近单周期稳定振荡;该方法用于多粒子模拟系统中,能较好改善束的品质,束晕-混沌现象能得到有效抑制.The Kapchinskij-Vladimirskij（K-V） distribution beam in the periodical focusing channels（PFCS） is selected as typical example,and a control method based on radial basis function neural network（RBFNN） is presented for control beam halo-chaos.The output of network as a controlling element is used to adjust focusing magnetic filed proportionally for control beam halo-chaos.Numerical results show that chaotic envelope radius of high-current beam can be controlled efficiently to the neighborhood of matched radius.This method is also applied to the multi-particle model.Under the control condition,the beam halos can be suppressed effectively,and quality of ion beam is improved evidently.This method provides a useful reference for controlling beam halo-chaos of high current beam in the PFCs.

关 键 词：强流离子束 束晕-混沌 混沌控制 RBF神经网络控制 

分 类 号：O415.50[理学—理论物理] TL501[理学—物理]