马尔可夫链在灌溉用水量预测中的应用

Application of Markov Chain in Prediction for Irrigation Water Consumption

依据漳河水库1963年~2008年灌溉用水量资料,应用均值标准差法建立4级分级标准。针对用水量为相依随机变量的特点,以各阶自相关系数为权重,运用马尔可夫链模型预测未来一年的灌溉用水量状态。结果表明,该方法直观、预测准确、计算简便,为灌区灌溉用水量的中长期预测提供了新的分析途径。

Based on the data of irrigation water consumption in Zhanghe Reservoir during 1963～2008,4 stages of classification criteria are set up by using a mean standard deviation method.Then,according to the characteristics with water consumption as random variable and with self-correlative coefficients in each order as weight,the Markov chain model is used for the prediction about the irrigation water consumption in the coming year.The rasults show that the method is direct,accurate and simple,so as to provide a new analyzing way for the mid and long term prediction of the irrigation water consumption in irrigation areas.