水利科技与经济
水利科技與經濟
수리과기여경제
2011年
12期
1-4
,共4页
组合预报%Bayes分类%加权Markov链%“马氏性”检验
組閤預報%Bayes分類%加權Markov鏈%“馬氏性”檢驗
조합예보%Bayes분류%가권Markov련%“마씨성”검험
combination forecast%Bayes classification%weighted Markov chain%＂Markov property＂ test
针对河川径流成因复杂性和水文过程随机性的特点,且用单一预测法存在一定局限性的现状,提出混合Bayes-Markov预测模型。先用Bayes公式对径流进行丰枯分类,然后采用加权Markov分析方法建立预测模型,该模型可综合利用Bayes和Markov方法的优点,提高径流预测精度。以兰州站河川径流量预测为例,进行模型验证。结果表明,2003～2009年径流量预测精度达到85.7%,能满足规范要求。
針對河川徑流成因複雜性和水文過程隨機性的特點,且用單一預測法存在一定跼限性的現狀,提齣混閤Bayes-Markov預測模型。先用Bayes公式對徑流進行豐枯分類,然後採用加權Markov分析方法建立預測模型,該模型可綜閤利用Bayes和Markov方法的優點,提高徑流預測精度。以蘭州站河川徑流量預測為例,進行模型驗證。結果錶明,2003～2009年徑流量預測精度達到85.7%,能滿足規範要求。
침대하천경류성인복잡성화수문과정수궤성적특점,차용단일예측법존재일정국한성적현상,제출혼합Bayes-Markov예측모형。선용Bayes공식대경류진행봉고분류,연후채용가권Markov분석방법건립예측모형,해모형가종합이용Bayes화Markov방법적우점,제고경류예측정도。이란주참하천경류량예측위례,진행모형험증。결과표명,2003～2009년경류량예측정도체도85.7%,능만족규범요구。
Aiming at the characteristics of the complexity of runoff cause and randomness of hydrological processes,and limitation of a single prediction method applied,a new method,called Bayes-Markov combined model,is presented based on Bayes theory and Markov theory.This paper attempts to use the Bayes formula to classify the low high annual runoff firstly,then to create forecasting model with the weighted Markov analysis method.The two prediction methods were scientifically combined,which generalizes advantages of the ones and raises the accuracy of runoff prediction.The prediction model was identified by taking prediction of annual runoff variation in Lanzhou Station of Yellow River Basin.The results show that the predicted values from 2003 to 2009 meet the requirements of the Specifications,and the accuracy of it was 85.7%.
