摘要
采用相空间重构理论计算实测月降雨的延迟时间、嵌入维数、G-P饱和关联维数和Laypunov指数,证明乌尔逊河流域月降雨时间序列存在混沌现象。使用LS-SVM预测模型和RBF神经网络预测模型,两种模型对乌尔逊河流域月降雨时间序列进行对比分析。在预测精度上,LS-SVM测模型的预测精度不太理想,而RBF神经网络预测模型在降雨量很少的月份精度也很低。若想在干旱区半干旱区的降雨预测中应用,需要进一步研究。
Using the phase space reconstruction calculated the real monthly rainfall＇s best delay time, embedded dimension, G-P saturated correlation dimension and Lyapunov exponent, and proving that chaos phenomena exists in the monthly rainfall time series of Wuerxun River Watershed. Comparing the results of monthly rainfall time series that there are used of the prediction model of Least squares support vector machine and radial basis function neural network model in Wuerxun River Watershed. However, in the accuracy, the accuracy of Least squares support vector machine is not ideal, and the accuracy of radial basis function neural network model is very low in rainfall rarely month. If the two models applied in the arid and semiarid regions to predicate rainfall, there are in need of further study.
出处
《水科学与工程技术》
2012年第2期1-4,共4页
Water Sciences and Engineering Technology
基金
国家水体污染控制与治理科技重大专项(2009ZX07106-006)
