摘要
检验太阳辐射时间序列是否有非线性特征,对于分析、建模和预测太阳辐射量是重要、有益的.提出用基于替代数据的检验方法来检验太阳辐射时间序列是否存在非线性特征,并将数据序列的三阶矩作为检验统计量.选取了美国Montana州Dillon地区和Wyoming州Green Rivet地区每日总辐射量、Utah州Moab地区的每月日平均总辐射量时间序列作为检验对象.数值分析的统计结果表明所研究的日总辐射时间序列存在非线性,而每月日平均总辐射时间序列未检测出非线性.因而,对太阳辐射时间序列建模和预测之前,检验其是否有非线性特征是必要的.
It is important and beneficial to test the nonlinearity of the solar radiation time series for their analysis, modeling and prediction. This paper tests the nonlinearity of the solar radiation time series based on the surrogate data method, and the third-order moment of the time series is used as the test statistic. The daily global solar radiation data from Dillon, Montana and Green River, Wyoming and the monthly average daily global solar radiation data are cosidered. The statistical results of the numerical analysis show that the daily global solar radiation time series exists nonliearity, whereas the monthly average daily global solar radiation time seires does not. So, it is necessary to test the nonliearity before modedling and predicting the solar radation.
出处
《数学的实践与认识》
CSCD
北大核心
2012年第22期137-142,共6页
Mathematics in Practice and Theory
基金
院级科研项目(2010C009)
合肥工业大学国家大学生创新项目
关键词
太阳辐射
非线性
替代数据
Solar radiation
nonlinearity
surrogate data
