基于优化组合核和Morlet小波核的LSSVM脉动风速预测方法

PDF(1356 KB)

振动与冲击 ›› 2016, Vol. 35 ›› Issue (18) : 52-57.

论文

迟恩楠，李春祥

作者信息 +

Forecasting fluctuating wind velocity using optimized combination kernel and Morlet wavelet kernel based LSSVM

Ennan Chi，Chunxiang Li

Author information +

文章历史 +

摘要

核函数是支持向量机的重要组成部分，直接影响预测模型的结果。根据Mercer定理，推导出了Morlet小波核函数，使其具有局部化、多层次、多分辨的优点。选择具有代表性的径向基（RBF）核函数和多项式（Poly）核函数构建出局部性和全局性相结合的线性组合核函数，使得预测模型保留RBF核函数所赋予的优越学习能力以及Poly核函数所拥有的强泛化能力。进一步，使用粒子群优化（PSO）算法，对惩罚参数、核参数、权重、尺度因子进行寻优，分别建立了基于Morlet小波核和组合核的PSO-LSSVM模型。使用建立的预测模型，对脉动风速进行了预测。通过比较预测性能评价指标，发现基于Morlet小波核和组合核PSO-LSSVM的预测精度优于常用的单核PSO-LSSVM模型。

Abstract

Kernel functions, which are important components of support vector machines (SVM), directly affect the results of prediction models. In accordance to the Mercer theorem, the present work has developed the Morlet wavelet kernel rendering the advantages of localization, multi-level and mufti-resolution. Evenmore, the representative radial basis function (RBF) kernel and polynomial (Poly) kernel functions are taken into consideration to construct a linear combination kernel function with both the local and global properties, thus forming the prediction models with both the superior learning ability and perfect generalization capability given by the RBF kernel and Poly kernel functions, respectively. Further, the particle swarm optimization (PSO) algorithm was used to optimize penalty parameter, kernel parameters, and weight and scale factor, then developing the PSO-LSSVM models using the Morlet wavelet kernel and combination kernel. By resorting to the proposed prediction models, the fluctuating wind velocity histories were forecasted. By comparing the predicting performance evaluation indices, it is found that the PSO-LSSVM models with the Morlet wavelet kernel and combination kernel functions renders more accurate results than the common single kernel (such as Poly and RBF) based PSO-LSSVM models.

导出引用

迟恩楠,李春祥 . 基于优化组合核和Morlet小波核的LSSVM脉动风速预测方法[J]. 振动与冲击, 2016, 35(18): 52-57

Ennan Chi，Chunxiang Li. Forecasting fluctuating wind velocity using optimized combination kernel and Morlet wavelet kernel based LSSVM[J]. Journal of Vibration and Shock, 2016, 35(18): 52-57

参考文献

[1] 申建红, 李春祥. 强风作用下超高层建筑风场特性的实测研究[J]. 振动与冲击, 2010, 29(5): 62-68.
   Shen Jianhong, Li Chunxiang. Researches on measurements of wind field characteristics of high-rise buildings under strong wind [J]. Journal of Vibration and Shock, 2010, 29(5): 62-68.
[2] 张华, 曾杰. 基于支持向量机的风速预测模型研究[J]. 太阳能学报，2010, 31(7): 928-931.
   Zhang Hua, Zeng Jie. Research on wind speed forecasting model based on support vector machine [J]. Journal of Solar Energy, 2010, 31(7): 928-931.
[3] 张广明, 袁宇浩, 龚松建. 基于改进最小二乘支持向量机方法的短期风速预测[J]. 上海交通大学报，2011, 45(8): 1125-1129.
   Zhang Guangming, Yuan Yuhao, Gong Songjian. Short term wind speed forecasting based on improved least square support vector machine method [J]. Journal of Shanghai Jiao Tong University, 2011, 45(8): 1125-1129.
[4] Rajasekaran S, Gayathri S, Lee T-L. Support vector regression methodology for storm surge predictions [J]. Ocean Engineering，2008, 35: 1578-1587.
[5] Chen WJ, Wang J. Application of support vector machine in industrial process [J]. Computers and Applied Chemistry, 2005, 22: 195-200.
[6] Suykens JAK, Vandewalle J. Least squares support vector machine classifiers [J]. Neural Processing Letters, 1999, 9: 293-300.
[7] Chen Thao-Tsen， Lee Shie-Jue. A weighted LS-SVM based learning system for time series forecasting [J]. Information Science，2015, 299: 99-116.
[8] 赵晨晖. 基于混合和函数支持向量机的基金投资决策研究[D]. 广东：华南理工大学，2012.
   Zhao Chenhui. The choice of funds based on support vector machine with mixed-kernel function [D]. Guangdong: South China University of Technology, 2012.
[9] Nourisola Hamid, Ahmadi Bahar. Robust adaptive H∞controller based on GA-Wavelet-SVM for nonlinear vehicle suspension with time delay actuator [J]. Journal of Vibration and Control，2015, 10: 1-10.
[10] Zhang Yin, Dai Miaolin, Ju Zhimin. Preliminary Discussion Regarding SVM Kernel Function Selection in the Twofold Rock Slope Prediction Model [J]. Journal of Computing in Civil Engineering，2015, 04015031: 1-8.
[11] Chen Fafa, Tang Baoping, Song Tao, Li Li. Multi-fault diagnosis study on roller bearing based on multi-kernel support vector machine with chaotic particle swarm optimization [J]. Measurement, 2014, 47: 576-590.
[12] 谷文成, 柴宝任, 滕艳平. 基于粒子群优化算法的支持向量机研究[J]. 北京理工大学报，2014, 34(7): 706-709.
   Gu Wencheng, Chai Baoren, Teng Yanping. Research on Support Vector Machine Based on Particle Swarm Optimization [J]. Transactions of Beijing Institute of Technology, 2014, 34(7): 706-709.