用于混沌时间序列预测的组合核函数最小二乘支持向量机

田中大; 高宪文; 石彤

doi:10.7498/aps.63.160508

摘要

针对混沌时间序列的预测问题，考虑到单一核函数的最小二乘支持向量机无法明显提高预测精度，提出了一种组合核函数的最小二乘支持向量机预测模型，模型中采用多项式函数与径向基函数组合构建核函数. 同时，还对遗传算法进行了改进，使之具有更快的收敛速度和更高的精度，改进的遗传算法适用于解决预测模型中的参数优化问题. 通过典型的Lorenz时间序列、Mackey-Glass时间序列、太阳黑子数时间序列以及具有混沌特性的网络流量时间序列对该模型进行了验证. 仿真结果表明所提出的模型是有效的.

关键词:

Abstract

Considering the problem that least squares support vector machine prediction model with single kernel function cannot significantly improve the prediction accuracy of chaotic time series, a combination kernel function least squares support vector machine prediction model is proposed. The model uses a polynomial function and radial basis function to construct the kernel function of least squares support vector machine. An improved genetic algorithm with better convergence speed and precision is proposed for parameter optimization of prediction model. The simulation experimental results of Lorenz, Mackey-Glass, Sunspot-Runoff in the Yellow River and chaotic network traffic time series demonstrate the effectiveness and characteristics of the proposed model.

Keywords:

作者及机构信息

1.
沈阳工业大学信息科学与工程学院, 沈阳 110870;

2.
东北大学信息科学与工程学院, 沈阳 110819;

3.
辽宁林业职业技术学院人文社会科学系, 沈阳 110101

基金项目: 国家自然科学基金重点项目（批准号：61034005）资助的课题.

Authors and contacts

1.
College of Information Science and Engineering, Shenyang University of Technology, Shenyang 110870, China;

2.
College of Information Science and Engineering, Northeastern University, Shenyang 110819, China;

3.
Department of Humanities, Liaoning Forestry Vocotion-technical College, Shenyang 110101, China

Funds: Project supported by the Key Program of the National Natural Science Foundation of China (Grant No. 61034005).

参考文献

[1]	Wang S W, Feng J, Liu G 2013 Math. Comput. Model. 58 677
[2]	Hong W C 2009 Energy Convers. Manage. 50 105
[3]	Zhao X J, Shang P J, Wang J 2013 Phys. Rev. E 87 805
[4]	Chowdhury P, Dwivedi B N 2011 Sol. Phys. 270 365
[5]	Ding J, Han L L, Chen X M 2010 Math. Comput. Model. 51 527
[6]	Toque C, Terraza V 2011 Commun. Statist.-Theor. Methods 40 1533
[7]	Khashei M, Bijari M 2011 Appl. Soft Comput. 11 2664
[8]	Li P, Li Y, Xiong Q, Chai Y, Zhang Y 2014 Int. J. Elec. Power 55 749
[9]	Li D C, Han M 2011 Acta Phys. Sin. 60 108903 (in Chinese) [李德才, 韩敏 2011 物理学报 60 108903]
[10]	Zhang J S, Xiao X C 2000 Chin. Phys. Lett. 17 88
[11]	Zhang J F, Hu S S 2007 Acta Phys. Sin. 56 713 (in Chinese) [张军峰, 胡寿松 2007 物理学报 56 713]
[12]	Zhang J S, Xiao X C 2001 Chin. Phys. 10 390
[13]	Li G D, Masuda S, Nagai M 2013 Int. J. Adv. Manuf. Tech. 67 2343
[14]	Li P, Tan Z, Yan L, Deng K 2011 Min. Sci. Tech. 21 557
[15]	Wang G L, Yang P C, Mao Y Q 2008 Acta Phys. Sin. 57 714 (in Chinese) [王革丽, 杨培才, 毛宇清 2008 物理学报 57 714]
[16]	Li H C, Zhang J S 2005 Chin. Phys. Lett. 22 2776
[17]	Vapnik V N 1999 IEEE Trans. Neural Netw. 10 1045
[18]	Tan W, Wang Y N, Zhou S W, Liu Z R 2003 Acta Phys. Sin. 52 2205 (in Chinese) [谭文, 王耀南, 周少武, 刘祖润 2003 物理学报 52 2205]
[19]	Yang Y F, Ren X M, Qin W Y, Wu Y F, Zhi X Z 2008 Acta Phys. Sin. 57 6139 (in Chinese) [杨永锋, 任兴民, 秦卫阳, 吴亚锋, 支希哲 2008 物理学报 57 6139]
[20]	Han M, Xu M L 2013 Acta Phys. Sin. 62 120510 (in Chinese) [韩敏, 许美玲 2013 物理学报 62 120510]
[21]	Song T, Li H 2012 Acta Phys. Sin. 61 080506 (in Chinese) [宋彤, 李菡 2012 物理学报 61 080506]
[22]	Bodyanskiy Y, Vynokurova O 2013 Inform. Sci. 220 170
[23]	Suykens J A K, Vandewalle J 1999 Neural Process Lett. 9 293
[24]	Pan J S, Hong M Z, Zhou Q F, Cai J Y, Wang H Z, Luo L K, Yang D Q, Dong J, Shi H X, Ren J L 2009 BMC Biotechnol. 9 52
[25]	Han X X, Xie G, Han X M, Xie K M 2012 Contr. Decis. 27 953 (in Chinese) [韩晓霞, 谢刚, 韩晓明, 谢克明 2012 控制与决策 27 953]
[26]	Smits G F, Jordaan E M 2002 Proceedings of the International Joint Conference on Neural Networks (Piscataway: IEEE) p2785
[27]	Wang S, Yang X J, Wei C J 2006 Proceedings of the World Congress on Intelligent Control and Automation (Piscataway: IEEE) p3014
[28]	Toledo C F M, Oliveira L, Franca P M 2014 J. Comput. Appl. Math. 261 341
[29]	Guo T, Lan J L, Li Y F, Jiang Y M 2013 J. Electron. Inform. Technol. 35 2220 (in Chinese) [郭通, 兰巨龙, 李玉峰, 江逸茗 2013 电子与信息学报 35 2220]
[30]	Meng Q F, Chen Y H, Feng Z Q, Wang F L, Chen S S 2013 Acta Phys. Sin. 62 150509 (in Chinese) [孟庆芳, 陈月辉, 冯志全, 王枫林, 陈珊珊 2013 物理学报 62 150509]
[31]	Cui W Z, Zhu C C, Bao W X, Liu J H 2005 Acta Phys. Sin. 54 3009 (in Chinese) [崔万照, 朱长纯, 保文星, 刘君华 2005 物理学报 54 3009]

施引文献

[1]	Wang S W, Feng J, Liu G 2013 Math. Comput. Model. 58 677
[2]	Hong W C 2009 Energy Convers. Manage. 50 105
[3]	Zhao X J, Shang P J, Wang J 2013 Phys. Rev. E 87 805
[4]	Chowdhury P, Dwivedi B N 2011 Sol. Phys. 270 365
[5]	Ding J, Han L L, Chen X M 2010 Math. Comput. Model. 51 527
[6]	Toque C, Terraza V 2011 Commun. Statist.-Theor. Methods 40 1533
[7]	Khashei M, Bijari M 2011 Appl. Soft Comput. 11 2664
[8]	Li P, Li Y, Xiong Q, Chai Y, Zhang Y 2014 Int. J. Elec. Power 55 749
[9]	Li D C, Han M 2011 Acta Phys. Sin. 60 108903 (in Chinese) [李德才, 韩敏 2011 物理学报 60 108903]
[10]	Zhang J S, Xiao X C 2000 Chin. Phys. Lett. 17 88
[11]	Zhang J F, Hu S S 2007 Acta Phys. Sin. 56 713 (in Chinese) [张军峰, 胡寿松 2007 物理学报 56 713]
[12]	Zhang J S, Xiao X C 2001 Chin. Phys. 10 390
[13]	Li G D, Masuda S, Nagai M 2013 Int. J. Adv. Manuf. Tech. 67 2343
[14]	Li P, Tan Z, Yan L, Deng K 2011 Min. Sci. Tech. 21 557
[15]	Wang G L, Yang P C, Mao Y Q 2008 Acta Phys. Sin. 57 714 (in Chinese) [王革丽, 杨培才, 毛宇清 2008 物理学报 57 714]
[16]	Li H C, Zhang J S 2005 Chin. Phys. Lett. 22 2776
[17]	Vapnik V N 1999 IEEE Trans. Neural Netw. 10 1045
[18]	Tan W, Wang Y N, Zhou S W, Liu Z R 2003 Acta Phys. Sin. 52 2205 (in Chinese) [谭文, 王耀南, 周少武, 刘祖润 2003 物理学报 52 2205]
[19]	Yang Y F, Ren X M, Qin W Y, Wu Y F, Zhi X Z 2008 Acta Phys. Sin. 57 6139 (in Chinese) [杨永锋, 任兴民, 秦卫阳, 吴亚锋, 支希哲 2008 物理学报 57 6139]
[20]	Han M, Xu M L 2013 Acta Phys. Sin. 62 120510 (in Chinese) [韩敏, 许美玲 2013 物理学报 62 120510]
[21]	Song T, Li H 2012 Acta Phys. Sin. 61 080506 (in Chinese) [宋彤, 李菡 2012 物理学报 61 080506]
[22]	Bodyanskiy Y, Vynokurova O 2013 Inform. Sci. 220 170
[23]	Suykens J A K, Vandewalle J 1999 Neural Process Lett. 9 293
[24]	Pan J S, Hong M Z, Zhou Q F, Cai J Y, Wang H Z, Luo L K, Yang D Q, Dong J, Shi H X, Ren J L 2009 BMC Biotechnol. 9 52
[25]	Han X X, Xie G, Han X M, Xie K M 2012 Contr. Decis. 27 953 (in Chinese) [韩晓霞, 谢刚, 韩晓明, 谢克明 2012 控制与决策 27 953]
[26]	Smits G F, Jordaan E M 2002 Proceedings of the International Joint Conference on Neural Networks (Piscataway: IEEE) p2785
[27]	Wang S, Yang X J, Wei C J 2006 Proceedings of the World Congress on Intelligent Control and Automation (Piscataway: IEEE) p3014
[28]	Toledo C F M, Oliveira L, Franca P M 2014 J. Comput. Appl. Math. 261 341
[29]	Guo T, Lan J L, Li Y F, Jiang Y M 2013 J. Electron. Inform. Technol. 35 2220 (in Chinese) [郭通, 兰巨龙, 李玉峰, 江逸茗 2013 电子与信息学报 35 2220]
[30]	Meng Q F, Chen Y H, Feng Z Q, Wang F L, Chen S S 2013 Acta Phys. Sin. 62 150509 (in Chinese) [孟庆芳, 陈月辉, 冯志全, 王枫林, 陈珊珊 2013 物理学报 62 150509]
[31]	Cui W Z, Zhu C C, Bao W X, Liu J H 2005 Acta Phys. Sin. 54 3009 (in Chinese) [崔万照, 朱长纯, 保文星, 刘君华 2005 物理学报 54 3009]

计量

文章访问数: 6253
PDF下载量: 839
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