文献类型：期刊文献

中文题名：时间序列模型在降水量预测中的应用研究

英文题名：Research on Precipitation Prediction Based on Time Series Model

作者：常青[1];赵晓莉[1]

第一作者：常青

机构：[1]新乡学院计算机与信息工程学院

第一机构：新乡学院计算机与信息工程学院

年份：2011

卷号：28

期号：7

起止页码：204-206

中文期刊名：计算机仿真

外文期刊名：Computer Simulation

收录：CSTPCD;;北大核心:【北大核心2008】；CSCD:【CSCD_E2011_2012】；

语种：中文

中文关键词：降水量预测;时间序列模型;支持向量机;多元回归

外文关键词：Precipitation prediction; Time series model; SVM; Multiple regressions

摘要：研究准确预测降水量,可提高应对灾害的能力。降水量的变化既受大气环流、地形、气压、气候带等各种环境因子的影响,降水量的动态特征呈现复杂非线性,使得准确预测未来降水量的变化较为困难。为了提高预测精度,采用融合时间序列模型与支持向量回归提出了一种新的多因子影响降水量预测模型。首先用支持向量机进行环境因子的非线性选择,用时间序列模型进行模型阶数的确定,最后以最优阶模型一步预测法检验模型外推能力。应用于赤峰地区夏季降水量预测,仿真结果表明,改进方法预测精度高,用在旱涝预测方面具有较好的应用前景。
The forecasting accuracy should be improved in the study on precipitation prediction.It is difficult to predict climate because of the dynamic characteristics of sample set as well as the effect of environmental factors.In order to improve the accuracy,a novel model based on time series and environmental factors was introduced in this paper.Firstly,the environmental factors were nonlinearly screened by support vector machine（SVM）.Secondly,the order was estimated by controlled autoregressive（CAR）.Lastly,reliability of SVM-CAR was validated by one-step prediction method.The simulation result of precipitation forecasting showed that this method has the advantages of high-precision and good prospect in drought and flood forecasting.