摘要
本文针对经典多层递阶方法所存在的缺陷,提出了一种改进方案—带有周期分量的多层递阶预报模型,它利用时间序列的显著周期分量取代原模型中的自回归部分,使之既考虑到各种物理因子的主导作用,又较好地反映了气象要素自身的周期性变化规律,因而预报效果较为稳定。
The defect of classical recursive estimate method and a improved for ecasting scheme-the recursive estimate forecasting model containing period component are presented. In this new model, the significance period component of time series has already replaced the autoregression component in classical recursive estimate model. Its forecasting effect is more stable than classical model.
出处
《成都信息工程学院学报》
1989年第2期41-48,共8页
Journal of Chengdu University of Information Technology
关键词
天气学
长期天气预报
周期分析
多层递阶方法
Synoptic meteorology
Long-range weather forecast
Period analysis
Recursive estimate method.
