马尔可夫链模型及其在旱涝预测的应用  

Markov Chain Model and Its Application in Drought and Flood Forecasting

作　　者：农吉夫[1] 

机构地区：[1]广西民族大学数学与计算机科学学院,广西南宁530006

出　　处：《广西民族大学学报（自然科学版）》2010年第1期70-75,共6页Journal of Guangxi Minzu University ：Natural Science Edition

摘　　要：利用样本均值—标准差方法将广西全区1958~2003年5月降水量分成丰涝、偏丰、偏旱、干旱4种状态,利用加权马尔可夫链预测方法对广西全区未来年份5月降水状态进行预测,并运用马尔可夫链性质得到了各降水状态发生的极限概率.结果表明,这种预测方法客观、准确,是区域降水量短期预测的科学有效的新途径.The mean and standard deviation of information series are taken as the classification standard of precipitation states,the precipitation numbers from 1958 to 2003 in Guangxi are classified into four classes according to the precipitation data,i.e.water logging year,weak water logging year,weak drought year,and drought year.The weighted Markov chain is applied to predict the precipitation grade in flood Season in the future years and the limited probability of precipitation state may be obtained according to the properties of Markov chain.The results show that the weighted Markov chain method is accurate in conclusion and simple in calculation,and it provides a new way to predict the state of regional precipitation.

关 键 词：马尔可夫链 降雨量 预测 

分 类 号：O211.62[理学—概率论与数理统计]