摘要
文章分别阐述了时间序列分析方法及Monte Carlo(蒙特卡洛)方法的基本原理和计算方法,并将其应用于降水预报的实例之中。时间序列分析是对降水资料进行观察、研究,用一定的数学方法建立预测模型,使时间序列向外延伸,从而获得序列的发展变化趋势,确定变量的预测值,精确的预报每一时段的降水量。蒙特卡洛方法是根据降水资料,提取研究区年降水量变化所蕴含的随机性和统计规律性,建立伪随机数与样品值的对应关系,从而对未来降水量进行预报,结果反映了预报时间内降水的总体特征,而不能精确到具体时间段。将这两种不同的方法用于同一地区年降水量的预报,对比结果,探其原因,分析其各自特点,进而得到更精确的预报值,对该地区水资源的研究有着重要的现实意义。
Basic principles,calculation methods and practical application of time series analysis and Monte Carlo method were described.With observing and studying precipitation data,time series analysis establishes the prediction model in certain mathematical methods to obtain the development trend of sequence and determine the predictive value of variables which can predict accurate precipitation for each period,while Monte Carlo extracts randomness and statistical regularity that contained in the change of annual precipitation.These attributes characterize the inherent characteristics of annual precipitation in the region,which can be used to establish correspondence between the sample values and pseudo-random number and to predict future rainfall,which just reflects the general characteristics.The two methods were used in the same area and results were consistent,which had important practical significance for water resource studies in this area.
出处
《环境科学与技术》
CAS
CSCD
北大核心
2011年第5期108-112,共5页
Environmental Science & Technology
基金
国家自然科学基金项目资助(41072171)
国家高技术研究发展计划863项目(2008AA06A410)
关键词
时间序列分析
蒙特卡洛
年降水量
预报
内蒙古五原县
time series analysis
Monte Carlo
annual rainfall
forecast
Wuyuan County
Inner Mongolia
作者简介
杨磊磊(1988-),女,在读硕士,主要从事水生态水环境及数值模拟研究,(手机)13620788432(电子信箱)yfdngleilei@yahoo_cn;
*通讯作者,(电话)0431-88502610(电子信箱)Luwenxiqlu.edu.cn。