摘要
提出了一种随机组合预测模型: 利用Mallat算法对水文时间序列进行多尺度分解, 得到对应尺度下的概貌(低频)分量和细节(高频)分量; 分别对概貌分量和细节分量建立随机模型进行预测, 预测结果的叠加即为原水文变量的预测。将该模型用于黄河三门峡站年径流预测中, 并与传统预测模型进行了对比分析, 结果表明, 建立的组合模型充分利用了现有信息, 预测精度高。
A hybrid stochastic model has been presented. First, the hydrologic time series are resolved into the detail sub-time series with high frequencyand outline ones with low frequencyin special scales by using Mallatalgorithm. Then, the stochastic model has been constructed for the resolved sub-time series. Finally, the superposition of predicted result is the predicted values of original time series.A case study of annual runoff at Sanmenxia station in the Yellow River is given. The results show that the suggested hybrid model can make full use of the information and the predicted accuracy has higher than the traditionalmodel.
出处
《高原气象》
CSCD
北大核心
2004年第z1期146-149,共4页
Plateau Meteorology
基金
国家自然科学基金项目(5027923)资助
关键词
小波分析
MALLAT算法
组合随机模型
年径流预测
Wavelet analysis
Mallat algorithm
Hybrid stochastic model
Prediction of annual runoff
