论文部分内容阅读
混沌时间序列预测是非线性动力学研究中一个十分重要的问题,支持向量回归方法为其提供了一种有效的解决思路.通过分析新样本加入训练集后支持向量集的变化情况,建立了一种混沌时间序列预测的支持向量回归算法,具备了在线学习的特点.同时,针对混沌信号提出了一种满足小波框架的小波核函数,它不但能以较高的精度逼近任意函数,而且适合于混沌信号的局部分析,提高了支持向量回归的泛化能力.最后就Mackey-Glass混沌时间序列在线预测问题进行了大量仿真.结果表明,本文算法与现有的算法相比具有训练时间短、预测精度高等特点,有一定的理论及实用价值.
Prediction of chaotic time series is a very important issue in the nonlinear dynamics research, and the support vector regression method provides an effective solution to the problem.After analyzing the changes of the support vector sets after the new samples are added to the training set, a new The chaotic time series prediction support vector regression algorithm possesses the characteristics of online learning.At the same time, a wavelet kernel function satisfying wavelet framework is proposed for chaotic signals, which not only can approximate any function with high precision, but also is suitable for chaos The local analysis of the signal improves the generalization ability of the support vector regression.Finally, a large number of simulations are carried out on the Mackey-Glass chaotic time series online prediction.The results show that compared with the existing algorithms, the proposed algorithm has the advantages of short training time, Higher characteristics, there is a certain theoretical and practical value.