欢迎光临,现在是     设为首页   加入收藏   联系我们
文章摘要
王小洁,陈珍莉,王旭,施晨晓,刘霄燕.基于相似日和Elman神经网络的逐时太阳总辐射预测研究[J].海南大学学报编辑部:自然科学版,2020,38(4):.
基于相似日和Elman神经网络的逐时太阳总辐射预测研究
A Study of Hourly Total Solar Radiation Prediction Based on Similar Day and Elman Neural Network
投稿时间:2020-06-04  修订日期:2020-08-31
DOI:
中文关键词: 太阳总辐射;灰色关联法;相似日;Elman神经网络
英文关键词: Global solar radiation; Grey correlation method; Similarity day; Elman neural network
基金项目:(HNQXQN201801);
作者单位E-mail
王小洁 海南省气象信息中心 wang_xj_work@163.com 
陈珍莉 海南省南海气象防灾减灾重点实验室
海南省气象信息中心 
 
王旭 海南省南海气象防灾减灾重点实验室
海南省气象信息中心 
 
施晨晓 海南省南海气象防灾减灾重点实验室
海南省气象信息中心 
 
刘霄燕 海南省南海气象防灾减灾重点实验室
海南省气象信息中心 
 
摘要点击次数: 57
全文下载次数: 0
中文摘要:
      采用Elman神经网络模型预估海口逐时太阳总辐射,并针对神经网络算法上存在泛化能力弱,提高训练精度困难,训练时间长等问题,将灰色关联分析法引入太阳总辐射预测模型旨在构建拟合精度优的模型。利用2009-2019年海口气象站逐时太阳总辐射曝辐量数据以及影响总辐射的气压、气温、相对湿度、降水量、日照等气象数据,分别构建基于Elman神经网络和基于相似日的Elman神经网络逐时太阳总辐射预测模型,其中2009-2018年为测试样本集,2019年数据作为检验数据集。并利用空间插值方法研究琼北地区辐射时空分布特征。结果表明,经过相似日筛选的辐射预测模型与观测值相关系数达0.97,平均相对误差在±0.5 MJ·m-2范围内,模型预测精度优于Elman神经网络算法。琼北总辐射整体空间分布均匀,具体呈现南多北少特征。时间分布上总辐射呈现冬季低夏季高,春秋季居中的特征。
英文摘要:
      The Elman neural network model is used to predict the hourly global solar radiation in Haikou, and for the shortcomings of the neural network algorithm such as difficulty in the ability of generalization, training approximation and long training time, a solar total radiation prediction model combined with the gray correlation analysis method was established to solve the problem.Hourly irradiation exposure of global radiation data at the Haikou Weather Station in 2009-2019, and meteorological data such as atmospheric pressure, temperature, relative humidity, precipitation, sunshine that affect the global radiation, the hourly solar radiation prediction models based on Elman neural network with or without similar day are constructed respectively, which the data in 2009-2018 is the training data set,and the data in 2019 is used as the validation data set. And a study of the temporal and spatial distribution characteristics of radiation in Qiongbei area by using the spatial interpolation method. The results show that the correlation coefficient of the global radiation prediction model and the observation value after the similar day screening reaches 0.97,the average relative error is within ±0.5 MJ·m-2, and the accuracy of this prediction model is better than the Elman neural network algorithm. The temporal distribution of global radiation in Qiongbei area was less in winter and the highest in summer, and centered in spring and autumn.the spatial distribution was even, showing the characteristics of more south and less north.
关闭