摘要
径流时间序列中包含了大量的信息,从频率域上来看,这些信息包括变化平稳的低频信息和变化相对剧烈的高频信息。本文利用小波分解,将径流时间序列分解为低频项和高频项,低频项采用逐步回归法预报,高频项采用基于自组织法求解的Volterra滤波器预报,两者结果综合,最终实现径流预报。实例计算表明,该模型具有足够高的计算精度。
A great deal of information is included in runoff time series. From the point of view of frequency, the information can be sorted into low frequency and high frequency items by wavelet decomposition. The stepwise regression algorithm and the Volterra filter based on group method of data handling （GMDH） are used to forecast them respectively. The results are integrated to forecast the runoff. At last the example of application shows that this model is of high accuracy.
出处
《水力发电学报》
EI
CSCD
北大核心
2006年第5期17-21,共5页
Journal of Hydroelectric Engineering
基金
国家自然科学基金资助(50279046)
