针对马尔科夫链模型在水文预测中的不足，提出基于权的马尔科夫模型。在传统马尔科夫链模型的基础上引入加权的概念，采用样本均方差方法将河流流量分成不同级别，以规范化的各阶自相关系数为权重，采用马尔可夫链的一步转移概率矩阵，以前期不同时段的状态作为初始状态，以其相应的转移概率矩阵的行向量作为预测时段指标值的状态概率，将同一状态的各预测概率的加权和作为处于该状态的预测概率，来预测和分析未来河流流量的状态，并以预测状态作为防洪规划的信息。文中以浑河沈阳站1960―2008年月均流量资料为例，对浑河沈阳城区段2015―2025年流量状态进行了预测，分析了出现丰水状态的概率和重现期，为浑河沈阳城区段防洪规划提供科学依据。%Based on the shortage of Markov chain model in hydrological forecasting,a method,which using the weighted Markov chains is developed.The introduction of the concept of weighted based on the traditional model of the Markov chain,using the sample variance method of river flow into different levels,by using stan-dardized self coefficients as weights ,using the one-step transition probability matrix of the Markov chain ,in different periods of early state as the initial state ,the state probability prediction index value of the period cor-responding to the row vectors of transition probability matrix as will ,the weighted prediction probability of the same state,and as a prediction probability of the state,to predict and analyze the future state of river flow,and to predict the state as flood forecast information.Taking Shenyang hydrological Station 1960~2008 average monthly river flow data as an example,the flow of Hun River in Shenyang from 2015 to 2025 are predicted and the probability of occurrence of water status and the return period are analyzed;it provide a scientific ba-sis for the prediction of flood at Hun River in Shenyang city.
展开▼
