微型电脑应用2026，Vol.42Issue(3)：240-243,4.
基于多源数据模态分解算法的气象短时预测算法
Meteorological Short-term Prediction Algorithm Based on Multi-source Data Mode Decomposition Algorithm
韩宏亮 1李明玥 2林嘉楠 1乔梁1
作者信息
- 1. 黑龙江省气象数据中心(黑龙江省气象探测中心、黑龙江省气象档案馆),黑龙江,哈尔滨 150030
- 2. 黑龙江省气候中心(黑龙江省气候变化中心),黑龙江,哈尔滨 150030
- 折叠
摘要
Abstract
Meteorological phenomena are affected by a variety of physical processes,and the interaction between these processes is complex,which makes the meteorological data show nonlinear and non-stationary characteristics.It is difficult to capture the high-frequency instantaneous changes and low-frequency periodic changes,resulting in the low accuracy of meteorological short-term prediction.Therefore,a meteorological short-term prediction algorithm based on multi-source data mode decomposition algorithm is proposed.The complete ensemble empirical mode decomposition with adaptive noise is used to decompose the multi-source meteorological time series data once to determine the high-frequency signal and low-frequency signal.The most complex high-frequency signal is decomposed twice by using variational mode decomposition to generate several eigenmode func-tion components,so as to obtain the high-frequency instantaneous changes and low-frequency periodic changes.The deep deter-ministic cyclic jump state network is constructed based on the obtained eigenmode function components,and the prediction re-sults of each network are superimposed to achieve short-term accurate prediction of meteorological.The experimental results show that the proposed algorithm has low Theil inequality coefficient,high protocol index,good performance of Taylor chart,and accurate meteorological short-term prediction results.关键词
多源数据/卡尔曼滤波/经验模态分解/气象短时预测Key words
multi-source data/Kalman filtering/empirical mode decomposition/meteorological short-term prediction分类
信息技术与安全科学引用本文复制引用
韩宏亮,李明玥,林嘉楠,乔梁..基于多源数据模态分解算法的气象短时预测算法[J].微型电脑应用,2026,42(3):240-243,4.
