###
气象:2014,40(7):881-885
本文二维码信息
码上扫一扫!
遗忘因子自适应最小二乘算法及其在气温预报中的应用
(1.北京应用气象研究所,北京 100029;2.中国人民解放军61741部队,北京 100094)
Forgetting Factor Adaptive Least Square Algorithm and Its Application in Temperature Forecasting
(1.Beijing Institute of Applied Meteorology, Beijing 100029;2.61741 Troops of People’s Liberation Army, Beijing 100094)
摘要
图/表
参考文献
相似文献
本文已被:浏览 1246次   下载 2193
投稿时间:2013-07-18    修订日期:2013-09-16
中文摘要: 海量数据的利用是建立自适应预报模型的基础,但随着数据的不断增加,新引入数据的作用会逐渐降低,有可能导致预报模型失效。为克服因数据量增加引起的所谓“数据饱和”现象对天气预报效果的影响,本文给出了考虑遗忘因子的线性自适应最小二乘建模算法的原理和方法,并利用该算法进行了最高气温和最低气温预报试验。结果表明,考虑遗忘因子的线性自适应建模算法优于传统的线性自适应建模算法,加入遗忘因子可以避免产生“数据饱和”现象,适当地选择遗忘因子有助于提高模型的预报准确率。
Abstract:A mass of data is the foundation of adaptive forecasting model. However, the role of new incoming data will be gradually reduced and the performance of the model will become poor with the data increasing. In order to overcome the influence of “data saturation” on the weather forecast, the method of adaptive linear least square modeling algorithm considering forgetting factors is developed and applied in max min temperature forecast. The results show that this adaptive linear least square modeling algorithm considering forgetting factors is superior to the traditional adaptive linear modeling algorithm, it can reduce the effect of “data saturation” by using the forgetting factor, and it is possible to improve the model’s forecast accuracy by choosing the appropriate forgetting factors.
文章编号:     中图分类号:    文献标志码:
基金项目:省部级2009年重点科研项目“概率天气预报业务系统”资助
引用文本:
翟宇梅,赵瑞星,高建春,王力维,韩海东,2014.遗忘因子自适应最小二乘算法及其在气温预报中的应用[J].气象,40(7):881-885.
ZHAI Yumei,ZHAO Ruixing,GAO Jianchun,WANG Liwei,HAN Haidong,2014.Forgetting Factor Adaptive Least Square Algorithm and Its Application in Temperature Forecasting[J].Meteor Mon,40(7):881-885.