首页
到顶部
到尾部
自动化毕业设计论文

Mackey-Glass 时间序列的模糊预测算法研究

分享到
时间:2013/1/8 22:03:40  作者:  来源:  查看:204  评论:0
内容摘要:Mackey-Glass 时间序列的模糊预测算法研究函数拟合是机器学习的重要研究内容,而以神经网络为代表的现代非经典数学方法在其中发挥了很大的优势,成为了目前机器学习的主要方法。本文首先对线性和非线性时间序列进行了研究;其次,以支持向量机复杂非线性系统为研究对象,将以非经典数学方...

Mackey-Glass 时间序列的模糊预测算法研究

函数拟合是机器学习的重要研究内容,而以神经网络为代表的现代非经典数学方法在其中发挥了很大的优势,成为了目前机器学习的主要方法。本文首先对线性和非线性时间序列进行了研究;其次,以支持向量机复杂非线性系统为研究对象,将以非经典数学方法引入非线性时间序列分析,采用支持向量机建立预测自适应模型。利用遗传算法构造了结构自适应非线时间序列预测模型,并用进行了详细验证;最后,针对神经网络模型泛化能力差和训练不确定的缺陷,本文引入了基于统计学习理论的机器学习方法——支持向量机,构造了支持向量机预测模型,研究并实现了支持向量机的SMO(Sequential Minimal Optimization)快速算法,同时分析了支持向量机的模型参数选取问题,运用遗传算法实现了SMO预测模型参数的优化;最后用国际标准数据和模拟仿真数据验证了预测模型的正确性。

 

关键词:非线性时间序列分析,预测,支持向量机,遗传算法

 

 


Non-linear time series forecasting model of support vector machine

ABSTRACT

Function regression is an important research content of machine learning, and non-classical mathematics represented by Artifical Neural Network (ANN) show great advantages to the issue, therefore they become the main techniques of machine learning. Firstly, This article linear and nonlinear time series have been studied .secondly, The complexity of support vector machines to nonlinear system is for the study, it will be the introduction of non-classical mathematical methods of nonlinear time series analysis, and use support vector machine model for the establishment of adaptive prediction. It structures of adaptive non-linear time series forecasting model for using the genetic algorithm, and carry out a detailed verification.; Lastly, because ANN has the limitation of the bad generalation and training uncertainty, the Support Vector Machine (SVM) which is based on Statistical Learning Theory (SLT) and has good generalization is introduced, and the issues are studied as follows: the forecasting model of SVM is constructed, Sequential Minimal Optimization (SMO) algorithm is studied  realized in order to improve the computation speed of SVM, the choice of SVM model parameters is analyzed, and then GA is utilized to optimize parameters of SMO forecasting model; the model is validated detailedly by international standard data, and the result shows the correctness of the model.

 

Key words:  Non-linear time series analysis, Forecasting, Support Vector Machine, Genetic Aigorithm

Tags:时间 函数 序列 模糊 预测 测算 算法 研究 机器 学习 重要 内容 神经 网络 

Tags:时间 序列 模糊 预测 测算 算法 研究   


客服QQ:496889925   电话15831080855
冀ICP备11001230号
Powered by OTCMS V2.3