摘要
混沌时间序列预测模型的参数对预测结果起着关键作用,传统上参数优化单独进行,忽略参数之间的联系,导致预测的精度比较低。为了提高混沌时间序列预测精度,提出一种基于遗传算法的混沌时间序列预测模型参数优化方法。参数优化方法的核心思想是相空间重构和支持向量机参数寻优同时进行,通过遗传算法算法的选择、交叉和变异操作达到参数优化求解问题。以经典混沌时间序列Mackey-G lass为例进行了验证性实验。实验结果表明,相对传统的参数寻优方法和分开优化的方法,方法时间复杂度低、预测精度高,是一种有效性的混沌时间序列预测模型参数优化算法。
Chaotic time series prediction model parameters play a key role in the predicted results Tradition methods separate the relationship between the two steps completely,which can not achieve the optimal performance for whole prediction model.A parameters hybrid optimization method is proposed based on genetic algorithm,which uses the interdependent relationship between phase space reconstruction and support vector machine parameters to improve the model performance.The main idea of the algorithm is searching support vector machines parameters during the phase space reconstructing by using genetic algorithm.The experiment results in sunspots and Mackey-Glass show that the proposed algorithm is more accurate and the computational complexity is lower compared with the traditional parameter optimization method and is an effective parameters optimization method of chaotic time series prediction model.
出处
《计算机仿真》
CSCD
北大核心
2011年第4期100-102,110,共4页
Computer Simulation
关键词
支持向量机
遗传算法
混沌时间序列
Support vector machine（SVM）
Genetic algorithm（GA）
Chaotic time series
