基于支持向量回归机的水文混沌时间序列预测

支持向量机在系统辨识和分类研究方面比较成熟,目前尚没有提出有效的支持向量回归理论来解决非线性、时变、干扰的复杂问题.支持向量回归机主要用于因果关系点对的回归预测,把支持向量回归机应用于水文混沌时间序列的预测研究是一个有意义的工作.在支持向量机一般理论基础上,提出了水文混沌时间序列支持向量回归机模型,并就模型进行仿真计算,讨论了模型参数对支持向量回归机预测精度的影响,为模型参数寻优提供一般指导原则.直门达水文站径流量混沌时间序列支持向量回归机预测实验表明,水文混沌时间序列支持向量回归机模型是有效的.

Prediction of Hydrologic Chaotic Time Series Based on Support Vector Regression

At present the support vector machine is mature in the system identification and the classified research,still had not proposed the support vector regression theory to solve complex system with non-linear,time-variable and disturbance.The support vector regression machine mainly uses to forecast relationship between cause and effect.According to the support vector machine theory,proposes a support vector regression model to hydrologic chaotic time series prediction,discusses to the influence of parameters on the model precision with simulation for guiding to choose the model parameters.The application of monthly runoff of Zhimenda indicates that the support vector regression model can deal with the complicated hydrologic data array well,and there is the good prediction precision.