搜索

x

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于分数阶最大相关熵算法的混沌时间序列预测

王世元 史春芬 钱国兵 王万里

引用本文:
Citation:

基于分数阶最大相关熵算法的混沌时间序列预测

王世元, 史春芬, 钱国兵, 王万里

Prediction of chaotic time series based on the fractional-order maximum correntropy criterion algorithm

Wang Shi-Yuan, Shi Chun-Fen, Qian Guo-Bing, Wang Wan-Li
PDF
导出引用
  • 为提高最大相关熵算法对混沌时间序列的预测速度和精度,提出了一种新的分数阶最大相关熵算法.在采用最大相关熵准则的基础上,利用分数阶微分设计了一种新的权重更新方法.在alpha噪声环境下,采用新的分数阶最大相关熵算法对Mackey-Glass和Lorenz两类具有代表性的混沌时间序列进行预测,并分析了分数阶的阶数对混沌时间序列预测性能的影响.仿真结果表明:与最小均方算法、最大相关熵算法以及分数阶最小均方算法三类自适应滤波算法相比,所提分数阶最大相关熵算法在混沌时间序列预测中能够有效地抑制非高斯脉冲噪声干扰的影响,具有较快收的敛速度和较低的稳态误差.
    Recently, adaptive filters have been widely used to perform the prediction of chaotic time series. Generally, the Gaussian noise is considered for the system noise. However, many non-Gaussian noises, e.g., impulse noise and alpha noise, exist in real systems. Adaptive filters are therefore required to reduce such non-Gaussian noises for practical applications. For improving the robustness against non-Gaussian noise, the maximum correntropy criterion (MCC) is successfully used to derive various robust adaptive filters. In these robust adaptive filters, the steepest ascent method based on the first-order derivative is generally utilized to construct the weight update form. It is well known that the traditional derivative can be generalized by the fractional-order derivative effectively. Therefore, to further improve the performance of adaptive filters based on the MCC, the fractional-order derivative is applied to the MCC-based algorithm, generating a novel fractional-order maximum correntropy criterion (FMCC) algorithm. Under the non-Gaussian noises, the proposed FMCC algorithm can be applied to predicting the chaotic time series effectively. In the proposed FMCC algorithm, the weight update form is constructed by using a combination of the first-order derivative based term and the fractional-order derivative based term. The Riemann-Liouville definition is utilized for calculating the fractional-order derivative in the proposed FMCC algorithm. The order of the fractional-order derivative is a crucial parameter of the proposed FMCC algorithm. However, it is difficult to obtain the optimal fractional order for different nonlinear systems theoretically. Therefore, the influence of the fractional order on the prediction performance is determined by trials for different nonlinear systems. The appropriate fractional order corresponds to the optimum of prediction accuracy, and can be chosen in advance. Simulations in the context of prediction of Mackey-Glass time series and Lorenz time series demonstrate that in the case of non-Gaussian noises the proposed FMCC algorithm achieves better prediction accuracy and faster convergence rate than the least mean square (LMS) algorithm, the MCC algorithm, and the fractional-order least mean square (FLMS) algorithm. In addition, the computational complexity of different filters is compared with each other under the example of the prediction of Marckey-Glass time series by using mean consumed time. It can be found that the computational complexity of FMCC algorithm is higher than those of the MCC and the LMS algorithms, but only slightly higher than that of the FLMS algorithm. As a result, comparing with other filters, the FMCC algorithm can improve the prediction performances of chaotic time series at the cost of the increasing computational complexity.
      通信作者: 王世元, wsy@swu.edu.cn
    • 基金项目: 国家自然科学基金(批准号:61671389)、中国博士后科学基金(批准号:2017M610583)、重庆市博士后科研项目特别资助(批准号:Xm2017107)和中央高校基本科研业务费(批准号:XDJK2017D177,XDJK2017D178)资助的课题.
      Corresponding author: Wang Shi-Yuan, wsy@swu.edu.cn
    • Funds: Project supported by the National Natural Science Foundation of China (Grant No. 61671389), China Postdoctoral Science Foundation Funded Project (Grant No. 2017M610583), Chongqing Postdoctoral Science Foundation Special Funded Project, China (Grant No. Xm2017107), and the Fundamental Research Funds for the Central Universities, China (Grant Nos. XDJK2017D177, XDJK2017D178).
    [1]

    Tang Z J, Ren F, Peng T, Wang W B 2014 Acta Phys. Sin. 63 050505(in Chinese) [唐舟进, 任峰, 彭涛, 王文博 2014 物理学报 63 050505]

    [2]

    Song T, Li H 2012 Acta Phys. Sin. 61 080506(in Chinese) [宋彤, 李菡 2012 物理学报 61 080506]

    [3]

    Zhang J S, Xiao X C 2000 Chin. Phys. Lett. 17 88

    [4]

    Farmer J D, Sidorowich J J 1987 Phys. Rev. Lett. 59 845

    [5]

    Zheng Y F, Wang S Y, Feng J C, Tse C K 2016 Digit. Signal Process. 48 130

    [6]

    Meng Q F, Zhang Q, Mou W Y 2006 Acta Phys. Sin. 55 1666(in Chinese) [孟庆芳, 张强, 牟文英 2006 物理学报 55 1666]

    [7]

    Takens F 1981 Lecture Notes Math. 898 366

    [8]

    Al-saggaf U M, Moinuddin M, Arif M, Zerguine A 2015 Signal Process. 111 50

    [9]

    Gui G, Peng W, Adachi F 2014 Int. J. Commun. Syst. 27 2956

    [10]

    Ozeki K, Umeda T 1984 Electr. Commun. Jpn. 67 19

    [11]

    Van V S, Lazarogredilla M, Santamaria I 2012 IEEE Trans. Neural Netw. Learn. Syst. 23 1313

    [12]

    Qiao B Q, Liu S M, Zeng H D, Li X, Dai B Z 2017 Sci. China: Phys. Mech. 60 040521

    [13]

    Erdogmus D, Principe J C 2002 IEEE Trans. Neural Netw. 13 1035

    [14]

    Hu T, Wu Q, Zhou D X 2016 IEEE Trans. Signal Process 64 6571

    [15]

    Chen B D, Xing L, Liang J L, Zheng N N, Principe J C 2014 Signal Process. Lett. 21 880

    [16]

    Shi L M, Lin Y 2014 Signal Process. Lett. 21 1385

    [17]

    Chen B D, Principe J C 2012 IEEE Trans. Process. Lett. 19 491

    [18]

    Chen Y, Li S G, Liu H 2016 Acta Phys. Sin. 65 170501(in Chinese) [陈晔, 李生刚, 刘恒 2016 物理学报 65 170501]

    [19]

    Shah S M, Samar R, Khan N M, Raja M A Z 2016 Nonlinear Dyn. 88 839

    [20]

    Zhou Y, Ionescu C, Machado J A T 2015 Nonlinear Dyn. 80 1661

    [21]

    Shah S M, Samar R, Raja M A Z, Chambers J A 2014 Electron. Lett. 50 973

    [22]

    Santamaria I, Pokharel P P, Principe J C 2006 IEEE Trans. Signal Process. 54 2187

    [23]

    Liu W, Pokharel P P, Principe J C 2007 IEEE Trans. Signal Process. 55 5286

    [24]

    Aronszajn A 1950 IEEE Trans. Am. Math. Soc. 68 337

    [25]

    Duan J W, Ding X, Liu T 2017 Sci. China: Inf. Sci. 60 1

    [26]

    Huang S, Zhang R, Chen D 2016 J. Computat. Nonlinear Dyn. 11 031007

    [27]

    Shoaib B, Qureshi I M 2014 Chin. Phys.. 23 050503

    [28]

    Mackey M C, Glass L 1977 Science 197 87

    [29]

    Lorenz E N 1963 J. Atmos. Sci. 20 130

    [30]

    Li B B, Ma H S, Liu M Q 2014 J. Electron. Inf. Technol. 36 868(in Chinese) [李兵兵, 马洪帅, 刘明骞 2014 电子与信息学报 36 868]

    [31]

    Stewart I 2000 Nature 406 948

  • [1]

    Tang Z J, Ren F, Peng T, Wang W B 2014 Acta Phys. Sin. 63 050505(in Chinese) [唐舟进, 任峰, 彭涛, 王文博 2014 物理学报 63 050505]

    [2]

    Song T, Li H 2012 Acta Phys. Sin. 61 080506(in Chinese) [宋彤, 李菡 2012 物理学报 61 080506]

    [3]

    Zhang J S, Xiao X C 2000 Chin. Phys. Lett. 17 88

    [4]

    Farmer J D, Sidorowich J J 1987 Phys. Rev. Lett. 59 845

    [5]

    Zheng Y F, Wang S Y, Feng J C, Tse C K 2016 Digit. Signal Process. 48 130

    [6]

    Meng Q F, Zhang Q, Mou W Y 2006 Acta Phys. Sin. 55 1666(in Chinese) [孟庆芳, 张强, 牟文英 2006 物理学报 55 1666]

    [7]

    Takens F 1981 Lecture Notes Math. 898 366

    [8]

    Al-saggaf U M, Moinuddin M, Arif M, Zerguine A 2015 Signal Process. 111 50

    [9]

    Gui G, Peng W, Adachi F 2014 Int. J. Commun. Syst. 27 2956

    [10]

    Ozeki K, Umeda T 1984 Electr. Commun. Jpn. 67 19

    [11]

    Van V S, Lazarogredilla M, Santamaria I 2012 IEEE Trans. Neural Netw. Learn. Syst. 23 1313

    [12]

    Qiao B Q, Liu S M, Zeng H D, Li X, Dai B Z 2017 Sci. China: Phys. Mech. 60 040521

    [13]

    Erdogmus D, Principe J C 2002 IEEE Trans. Neural Netw. 13 1035

    [14]

    Hu T, Wu Q, Zhou D X 2016 IEEE Trans. Signal Process 64 6571

    [15]

    Chen B D, Xing L, Liang J L, Zheng N N, Principe J C 2014 Signal Process. Lett. 21 880

    [16]

    Shi L M, Lin Y 2014 Signal Process. Lett. 21 1385

    [17]

    Chen B D, Principe J C 2012 IEEE Trans. Process. Lett. 19 491

    [18]

    Chen Y, Li S G, Liu H 2016 Acta Phys. Sin. 65 170501(in Chinese) [陈晔, 李生刚, 刘恒 2016 物理学报 65 170501]

    [19]

    Shah S M, Samar R, Khan N M, Raja M A Z 2016 Nonlinear Dyn. 88 839

    [20]

    Zhou Y, Ionescu C, Machado J A T 2015 Nonlinear Dyn. 80 1661

    [21]

    Shah S M, Samar R, Raja M A Z, Chambers J A 2014 Electron. Lett. 50 973

    [22]

    Santamaria I, Pokharel P P, Principe J C 2006 IEEE Trans. Signal Process. 54 2187

    [23]

    Liu W, Pokharel P P, Principe J C 2007 IEEE Trans. Signal Process. 55 5286

    [24]

    Aronszajn A 1950 IEEE Trans. Am. Math. Soc. 68 337

    [25]

    Duan J W, Ding X, Liu T 2017 Sci. China: Inf. Sci. 60 1

    [26]

    Huang S, Zhang R, Chen D 2016 J. Computat. Nonlinear Dyn. 11 031007

    [27]

    Shoaib B, Qureshi I M 2014 Chin. Phys.. 23 050503

    [28]

    Mackey M C, Glass L 1977 Science 197 87

    [29]

    Lorenz E N 1963 J. Atmos. Sci. 20 130

    [30]

    Li B B, Ma H S, Liu M Q 2014 J. Electron. Inf. Technol. 36 868(in Chinese) [李兵兵, 马洪帅, 刘明骞 2014 电子与信息学报 36 868]

    [31]

    Stewart I 2000 Nature 406 948

  • [1] 黄颖, 顾长贵, 杨会杰. 神经网络超参数优化的删除垃圾神经元策略. 物理学报, 2022, 71(16): 160501. doi: 10.7498/aps.71.20220436
    [2] 齐乐天, 王世元, 沈明琳, 黄刚毅. 基于Nyström柯西核共轭梯度算法的混沌时间序列预测. 物理学报, 2022, 71(10): 108401. doi: 10.7498/aps.71.20212274
    [3] 倪龙, 陈晓. 基于频散补偿和分数阶微分的多模式兰姆波分离. 物理学报, 2018, 67(20): 204301. doi: 10.7498/aps.67.20180561
    [4] 温少芳, 申永军, 杨绍普. 分数阶时滞反馈对Duffing振子动力学特性的影响. 物理学报, 2016, 65(9): 094502. doi: 10.7498/aps.65.094502
    [5] 韦鹏, 申永军, 杨绍普. 分数阶van der Pol振子的超谐共振. 物理学报, 2014, 63(1): 010503. doi: 10.7498/aps.63.010503
    [6] 唐舟进, 任峰, 彭涛, 王文博. 基于迭代误差补偿的混沌时间序列最小二乘支持向量机预测算法. 物理学报, 2014, 63(5): 050505. doi: 10.7498/aps.63.050505
    [7] 陈晓, 汪陈龙. 基于赛利斯模型和分数阶微分的兰姆波信号消噪. 物理学报, 2014, 63(18): 184301. doi: 10.7498/aps.63.184301
    [8] 唐舟进, 彭涛, 王文博. 一种基于相关分析的局域最小二乘支持向量机小尺度网络流量预测算法. 物理学报, 2014, 63(13): 130504. doi: 10.7498/aps.63.130504
    [9] 王新迎, 韩敏, 王亚楠. 含噪混沌时间序列预测误差分析. 物理学报, 2013, 62(5): 050504. doi: 10.7498/aps.62.050504
    [10] 申永军, 杨绍普, 邢海军. 含分数阶微分的线性单自由度振子的动力学分析. 物理学报, 2012, 61(11): 110505. doi: 10.7498/aps.61.110505
    [11] 王新迎, 韩敏. 基于极端学习机的多变量混沌时间序列预测. 物理学报, 2012, 61(8): 080507. doi: 10.7498/aps.61.080507
    [12] 申永军, 杨绍普, 邢海军. 含分数阶微分的线性单自由度振子的动力学分析(Ⅱ). 物理学报, 2012, 61(15): 150503. doi: 10.7498/aps.61.150503
    [13] 宋彤, 李菡. 基于小波回声状态网络的混沌时间序列预测. 物理学报, 2012, 61(8): 080506. doi: 10.7498/aps.61.080506
    [14] 宋青松, 冯祖仁, 李人厚. 用于混沌时间序列预测的多簇回响状态网络. 物理学报, 2009, 58(7): 5057-5064. doi: 10.7498/aps.58.5057
    [15] 张勇, 关伟. 基于最大Lyapunov指数的多变量混沌时间序列预测. 物理学报, 2009, 58(2): 756-763. doi: 10.7498/aps.58.756
    [16] 韩 敏, 史志伟, 郭 伟. 储备池状态空间重构与混沌时间序列预测. 物理学报, 2007, 56(1): 43-50. doi: 10.7498/aps.56.43
    [17] 胡玉霞, 高金峰. 一种预测混沌时间序列的模糊神经网络方法. 物理学报, 2005, 54(11): 5034-5038. doi: 10.7498/aps.54.5034
    [18] 甘建超, 肖先赐. 基于相空间邻域的混沌时间序列自适应预测滤波器(Ⅰ)线性自适应滤波. 物理学报, 2003, 52(5): 1096-1101. doi: 10.7498/aps.52.1096
    [19] 韦保林, 罗晓曙, 汪秉宏, 全宏俊, 郭维, 傅金阶. 一种基于三阶Volterra滤波器的混沌时间序列自适应预测方法. 物理学报, 2002, 51(10): 2205-2210. doi: 10.7498/aps.51.2205
    [20] 张家树, 肖先赐. 用于混沌时间序列自适应预测的一种少参数二阶Volterra滤波器. 物理学报, 2001, 50(7): 1248-1254. doi: 10.7498/aps.50.1248
计量
  • 文章访问数:  5198
  • PDF下载量:  367
  • 被引次数: 0
出版历程
  • 收稿日期:  2017-08-08
  • 修回日期:  2017-09-09
  • 刊出日期:  2018-01-05

/

返回文章
返回