| 一种多时间尺度SVM局部短时临近降雨预测方法 |
| |
| 引用本文： | 贺佳佳,陈凯,陈劲松,徐文文,唐历,刘军.一种多时间尺度SVM局部短时临近降雨预测方法[J].气象,2017,43(4):402-412. |
| |
| 作者姓名： | 贺佳佳 陈凯 陈劲松 徐文文 唐历 刘军 |
| |
| 作者单位： | 深圳市气象局，深圳 518040
深圳南方强天气研究重点实验室，深圳 518040,中国科学院深圳先进技术研究院，深圳 518055,中国科学院深圳先进技术研究院，深圳 518055,深圳市气象局，深圳 518040,深圳市国家气候观象台，深圳 518040,中国科学院深圳先进技术研究院，深圳 518055 |
| |
| 基金项目： | 深圳南方强天气研究重点实验室项目(ZDSYS20140715153957030和SZQX2015113)及广东省科技厅项目(2014A020218014)共同资助 |
| |
| 摘 要： | 近年来支持向量机(support vector machine,SVM)在气象领域得到了广泛应用,在该类应用中单一建模是目前普遍采用的一种思路,单一建模方法寻找的是大而泛的预测模型,预测的目标以面降雨为主。本研究针对每个气象站点进行单独动态建模,建模方法为多时间尺度SVM,探索建立一种动态SVM短时临近降水预测模型,充分考虑不同站点、不同时刻的气象要素差异,初步解决了单一建模过于注重整体规律、建立固定的整体预测函数模型而忽略不同站点、不同时刻局部气象变化的不足,并尝试提高短时临近降水预报的准确率。初步实现了地理空间上更高密度、更精细化的降雨预测,时间分辨率为1 h,TS评分始终保持在较高的水平,对1 h预测的TS评分平均可达40%以上,部分站点接近50%,且模型预测准确率具有一定的稳定性和参考价值。
|
| 关 键 词： | 短时临近降雨预测，支持向量机，时间尺度，TS检验 |
| 收稿时间： | 2016/9/14 0:00:00 |
| 修稿时间： | 2016/12/23 0:00:00 |
A Multi Time Scales SVM Method for Local Short Term Rainfall Prediction |
| |
| HE Jiaji,CHEN Kai,CHEN Jinsong,XU Wenwen,TANG Li and LIU Jun.A Multi Time Scales SVM Method for Local Short Term Rainfall Prediction[J].Meteorological Monthly,2017,43(4):402-412. |
| |
| Authors: | HE Jiaji CHEN Kai CHEN Jinsong XU Wenwen TANG Li and LIU Jun |
| |
| Abstract: | In recent years, SVM (support vector machine) has been widely used in meteorological field. Single modeling is the most common approach for this type of application which just looks for a large, gen eric prediction mode to forecast surface rainfall. In this study, individual meteorological stations were modeled dynamically through multi time scale SVM. So we established a dynamic short term rainfall forecasting model and fully considered the difference of meteorological elements at different time stamps of different sites, solving the problem that the single fixed global model is concerned with the whole law too much and neglects the difficiency of local meteorological changes at different sites and different times. Therefore, our method has the ability of improving the accuracy of short term precipitation forecast. In our study, the prediction for higher density and finer rainfall in geographical space was basically achieved, the temporal resolution was 1 h, and the TS score was always kept at a high level. As a result, the average TS score of 1 h forecast is more than 40%, and for some sites, it is close to 50%. Thus, the prediction accuracy of the model has certain stability and reference value. |
| |
| Keywords: | |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《气象》浏览原始摘要信息 |
| 点击此处可从《气象》下载免费的PDF全文 |
|
