迟滞混沌神经元/网络的控制策略及应用研究

修春波; 刘畅; 郭富慧; 成怡; 罗菁

doi:10.7498/aps.64.060504

摘要

为了保持神经网络在优化计算求解过程中结构不被改变, 以迟滞混沌神经元和迟滞混沌神经网络为研究对象, 提出了一种基于滤波跟踪误差的控制策略来实现神经元/网络的稳定控制. 采用该控制策略, 在不改变非线性特性发生机理的情况下, 神经元/网络可实现函数优化计算问题的求解. 所设计的控制律包含两部分: 一部分是系统进入滤波跟踪误差面时的等效控制部分, 另一部分为确保系统快速进入滤波跟踪误差面的控制部分. 采用Lyapunov方法对神经元/网络的控制进行了稳定性证明. 根据待寻优函数直接求得神经元的控制律, 在该控制律的作用下, 神经元/网络可逐渐稳定到优化函数的极值点, 从而实现优化问题的求解, 仿真实验结果验证了该控制方法在优化计算中的可行性和有效性.

关键词:

迟滞 /
混沌 /
神经元 /
神经网络

Abstract

In order to remain the structure of the neural network in the process of the optimization unchanged, taking the hysteretic chaotic neuron and the hysteretic chaotic neural network as controlled plants, a novel control strategy based on the filtered tracking error is proposed to perform the stability control for the single hysteretic chaotic neuron or the hysteretic chaotic neural network. Especially, the hysteretic chaotic neuron and the hysteretic chaotic neural network can be used to solve the optimization problem through using the control strategy on condition that the generation mechanisms of the nonlinear characteristics, hysteresis and chaos, are unchanged. The control law is composed of two terms: one is the equivalent control term in the ideal filtered tracking error surface, and the other is the control term which can make the system reach the filtered tracking error surface quickly. Lyapunov stability method is used to prove the stability of the control strategy for the single hysteretic chaotic neuron and hysteretic chaotic neural network. The control laws of hysteretic chaotic neurons can be obtained according to the optimization function. The state of the single hysteretic chaotic neuron or the hysteretic chaotic neural network can converge to an extreme point of the optimization function gradually by the control law. In this way, the optimization problem can be solved effectively. Simulation results prove the feasibility and validity of the control strategy for optimization problem.

Keywords:

hysteretic /
chaos /
neuron /
neural network

作者及机构信息

1.
天津工业大学, 电工电能新技术天津市重点实验室, 天津 300387;

2.
天津工业大学电气工程与自动化学院, 天津 300387

基金项目: 国家自然科学基金(批准号: 61203302)和天津市应用基础与前沿技术研究计划(批准号: 14JCYBJC18900)资助的课题.

Authors and contacts

1.
Key Laboratory of Advanced Electrical Engineering and Energy Technology, Tianjin Polytechnic University, Tianjin 300387, China;

2.
School of Electrical Engineering and Automation, Tianjin Polytechnic University, Tianjin 300387, China

Funds: Project supported by the National Natural Science Foundation of China (Grant No. 61203302) and the Tianjin Research Program of Application Foundation and Advanced Technology, China (Grant No. 14JCYBJC18900).

参考文献

[1]	Bosque G, Campo I D, Echanobe J 2014 Eng. Appl. Artif. Intel. 32 283
[2]	Liu X D, Xiu C B 2007 Neurocomputing 70 2561
[3]	Xia J W, Park J H, Zeng H B, Shen H 2014 Neurocomputing 140 210
[4]	Wang X, Li C D, Huang T W 2014 Neurocomputing 140 155
[5]	Lang J, Hao Z C 2014 Opt. Laser Eng. 52 91
[6]	Yu S J, Huan R S, Zhang Y, Feng D 2014 Acta Phys. Sin. 63 060701 (in Chinese) [于舒娟, 宦如松, 张昀, 冯迪 2014 63 060701]
[7]	Wang X Y, Bao X M 2013 Chin. Phys. B 22 050508
[8]	Kalpana M, Balasubramaniam P 2013 Chin. Phys. B 22 078401
[9]	Kwon O M, Park M J, Park J H, Lee S M, Cha E J 2013 Chin. Phys. B 22 110504
[10]	Zeng Z Z 2013 Acta Phys. Sin. 62 030504 (in Chinese) [曾喆昭 2013 62 030504]
[11]	Cao J D, Lu J 2006 Chaos 16 013133
[12]	Huang X, Cao J D 2006 Nonlinearity 19 2797
[13]	He W L, Cao J D 2009 Nonlinear Dynam. 55 55
[14]	Zhu Q X, Cao J D 2010 Nonlinear Dynam. 61 517
[15]	Cao J D, Alofi A, Al-Mazrooei A, Elaiw A 2013 Abstr. Appl. Anal. 2013 940573
[16]	Liu X D, Xiu C B 2008 Neural Comput. Appl. 17 579
[17]	Sun M, Zhao L, Ding J C, Zao X 2010 Syst. Eng. Electron. 32 396 (in Chinese) [孙明, 赵琳, 丁继成, 赵欣 2010 系统工程与电子技术 32 396]
[18]	Xiu C B, Liu Y X, Lu L F 2010 Control Eng. China 17 300 (in Chinese) [修春波, 刘玉霞, 陆丽芬 2010 控制工程 17 300]
[19]	Yang G, Yi J Y 2014 Neurocomputing 127 114
[20]	Ding Z, Leung H, Zhu Z W 2002 Math. Comput. Model. 36 1007
[21]	Zhang Q H Y, Xie X P, Zhu P, Chen H P, He G G 2014 Commun. Nonlinear Sci. 19 2793
[22]	Li X D, Song S J 2014 Commun. Nonlinear Sci. 19 3892
[23]	Zhang X D, Zhu P, Xie X P, He G G 2013 Acta Phys. Sin. 62 210506 (in Chinese) [张旭东, 朱萍, 谢小平, 何国光 2013 62 210506]
[24]	Jagannathan S, Vandegrift M W, Lewis F L 2000 Automatica 36 229
[25]	Jagannathan S, Lewis F L 2000 Inform. Sci. 123 223

施引文献

[1]	Bosque G, Campo I D, Echanobe J 2014 Eng. Appl. Artif. Intel. 32 283
[2]	Liu X D, Xiu C B 2007 Neurocomputing 70 2561
[3]	Xia J W, Park J H, Zeng H B, Shen H 2014 Neurocomputing 140 210
[4]	Wang X, Li C D, Huang T W 2014 Neurocomputing 140 155
[5]	Lang J, Hao Z C 2014 Opt. Laser Eng. 52 91
[6]	Yu S J, Huan R S, Zhang Y, Feng D 2014 Acta Phys. Sin. 63 060701 (in Chinese) [于舒娟, 宦如松, 张昀, 冯迪 2014 63 060701]
[7]	Wang X Y, Bao X M 2013 Chin. Phys. B 22 050508
[8]	Kalpana M, Balasubramaniam P 2013 Chin. Phys. B 22 078401
[9]	Kwon O M, Park M J, Park J H, Lee S M, Cha E J 2013 Chin. Phys. B 22 110504
[10]	Zeng Z Z 2013 Acta Phys. Sin. 62 030504 (in Chinese) [曾喆昭 2013 62 030504]
[11]	Cao J D, Lu J 2006 Chaos 16 013133
[12]	Huang X, Cao J D 2006 Nonlinearity 19 2797
[13]	He W L, Cao J D 2009 Nonlinear Dynam. 55 55
[14]	Zhu Q X, Cao J D 2010 Nonlinear Dynam. 61 517
[15]	Cao J D, Alofi A, Al-Mazrooei A, Elaiw A 2013 Abstr. Appl. Anal. 2013 940573
[16]	Liu X D, Xiu C B 2008 Neural Comput. Appl. 17 579
[17]	Sun M, Zhao L, Ding J C, Zao X 2010 Syst. Eng. Electron. 32 396 (in Chinese) [孙明, 赵琳, 丁继成, 赵欣 2010 系统工程与电子技术 32 396]
[18]	Xiu C B, Liu Y X, Lu L F 2010 Control Eng. China 17 300 (in Chinese) [修春波, 刘玉霞, 陆丽芬 2010 控制工程 17 300]
[19]	Yang G, Yi J Y 2014 Neurocomputing 127 114
[20]	Ding Z, Leung H, Zhu Z W 2002 Math. Comput. Model. 36 1007
[21]	Zhang Q H Y, Xie X P, Zhu P, Chen H P, He G G 2014 Commun. Nonlinear Sci. 19 2793
[22]	Li X D, Song S J 2014 Commun. Nonlinear Sci. 19 3892
[23]	Zhang X D, Zhu P, Xie X P, He G G 2013 Acta Phys. Sin. 62 210506 (in Chinese) [张旭东, 朱萍, 谢小平, 何国光 2013 62 210506]
[24]	Jagannathan S, Vandegrift M W, Lewis F L 2000 Automatica 36 229
[25]	Jagannathan S, Lewis F L 2000 Inform. Sci. 123 223

[1]	马少卿, 龚士香, 张微, 路承彪, 李小俚, 李英伟. 宽带微量太赫兹辐射促进神经元生长发育. , 2022, 71(20): 208701. doi: 10.7498/aps.71.20220636
[2]	张秀芳, 马军, 徐莹, 任国栋. 光电管耦合FitzHugh-Nagumo神经元的同步. , 2021, 70(9): 090502. doi: 10.7498/aps.70.20201953
[3]	于文婷, 张娟, 唐军. 动态突触、神经耦合与时间延迟对神经元发放的影响. , 2017, 66(20): 200201. doi: 10.7498/aps.66.200201
[4]	魏德志, 陈福集, 郑小雪. 基于混沌理论和改进径向基函数神经网络的网络舆情预测方法. , 2015, 64(11): 110503. doi: 10.7498/aps.64.110503
[5]	陈铁明, 蒋融融. 混沌映射和神经网络互扰的新型复合流密码. , 2013, 62(4): 040301. doi: 10.7498/aps.62.040301
[6]	王兴元, 任小丽, 张永雷. 参数未知神经元模型的全阶与降阶最优同步. , 2012, 61(6): 060508. doi: 10.7498/aps.61.060508
[7]	陈军, 李春光. 禁忌学习神经元模型的电路设计及其动力学研究. , 2011, 60(2): 020502. doi: 10.7498/aps.60.020502
[8]	李华青, 廖晓峰, 黄宏宇. 基于神经网络和滑模控制的不确定混沌系统同步. , 2011, 60(2): 020512. doi: 10.7498/aps.60.020512
[9]	王慧巧, 俞连春, 陈勇. 离子通道噪声对神经元新陈代谢能量的影响. , 2009, 58(7): 5070-5074. doi: 10.7498/aps.58.5070
[10]	乔晓艳, 李刚, 董有尔, 贺秉军. 弱激光诱导神经元兴奋性改变的实验研究. , 2008, 57(2): 1259-1265. doi: 10.7498/aps.57.1259
[11]	赵海全, 张家树. 混沌通信系统中非线性信道的自适应组合神经网络均衡. , 2008, 57(7): 3996-4006. doi: 10.7498/aps.57.3996
[12]	王永生, 孙瑾, 王昌金, 范洪达. 变参数混沌时间序列的神经网络预测研究. , 2008, 57(10): 6120-6131. doi: 10.7498/aps.57.6120
[13]	彭建华, 于洪洁. 神经系统中随机和混沌感知信号的联想记忆与分割. , 2007, 56(8): 4353-4360. doi: 10.7498/aps.56.4353
[14]	王瑞敏, 赵鸿. 神经元传输函数对人工神经网络动力学特性的影响. , 2007, 56(2): 730-739. doi: 10.7498/aps.56.730
[15]	乔晓艳, 李刚, 林凌, 贺秉军. 弱激光对神经元钾离子通道特性影响的实验研究. , 2007, 56(4): 2448-2455. doi: 10.7498/aps.56.2448
[16]	牛培峰, 张君, 关新平. 基于遗传算法的统一混沌系统比例-积分-微分神经网络解耦控制研究. , 2007, 56(5): 2493-2497. doi: 10.7498/aps.56.2493
[17]	行鸿彦, 徐伟. 混沌背景中微弱信号检测的神经网络方法. , 2007, 56(7): 3771-3776. doi: 10.7498/aps.56.3771
[18]	王耀南, 谭文. 混沌系统的遗传神经网络控制. , 2003, 52(11): 2723-2728. doi: 10.7498/aps.52.2723
[19]	谭文, 王耀南, 刘祖润, 周少武. 非线性系统混沌运动的神经网络控制. , 2002, 51(11): 2463-2466. doi: 10.7498/aps.51.2463
[20]	关新平, 唐英干, 范正平, 王益群. 基于神经网络的混沌系统鲁棒自适应同步. , 2001, 50(11): 2112-2115. doi: 10.7498/aps.50.2112

计量

文章访问数: 7000
PDF下载量: 261
被引次数: 0

姓名
邮箱
手机号码
标题
留言内容
验证码

搜索

留言板