基于降水-径流模型的中长期径流预测

摘要

Through the analysis made on the physical causation, the correlation between precipitation and runoff is determined herein, and furthermore, a significant correlation between precipitation and runoff is proved at Guide Hydrological Station on the Upper Yellow River with the related probabilistic approach through chi-square test On this basis, a precipitation-runoff forecasting model is set up based on the least squares support vector machines. The calculation result shows that within the predicted values of ten points, all the predicted flow-rate values of them are less than 10% compared with the mean annual value concerned, except only one that cannot meet the requirement of less than 10% for the flow rate departure, therefore, a better prediction accuracy can be obtained as well%通过物理成因分析确定了降水与径流的相关性,并采用相关概率法,利用卡方检验证明黄河上游贵德站降水与径流相关关系显著.在此基础上,采用最小二乘支持向量机方法建立了降水-径流预测模型.计算结果表明,预测的十个点位值,除一个不满足流量距平值小于10％的要求以外,余者的预测流量值与多年平均值相比均小于10％,预测精度相对较好.