摘要
以柳林泉域泉流量及降水量的系列资料为基础,分别建立了基于灰色理论的GM(1,1)模型、GM(1,1)残差周期修正模型和不同方案下的多元线性回归模型,模拟预测了泉流量动态,并对泉流量衰减原因进行了分析。结果表明:残差周期修正后的GM(1,1)模型优于多元线性回归模型;还原泉流量下的回归模型较实测泉流量下的回归模型拟合误差大但预测误差小;柳林泉流量的衰减过程分为两个阶段,1990年以前大气降水减少是泉流量衰减的主要原因,1990年以后泉流量的衰减则是大气降水减少和人工开采量增大共同作用所致。
On the basis of data series on spring discharge and precipitation of Liulin spring area, the GM （ 1,1 ） model, period residual modification of GM （ 1,1 ） model and multiple linear regression model in different schemes were established respectively based on grey system theory to simulate and predict the spring discharge. The attenuation causes of flow rate were analyzed as well. The results show that the period residual modification of GM （ 1,1 ） model is superior to the multiple linear regression model. Compare fitting precision of spring flow concluded from multiple linear re- gression model to from actual measurement state, the former is lower but forecast error is lower also. The attenuation process of spring flux is divided into two periods： before 1990, decreasing precipitation is main factor for attenuation, and after then, both of decreasing precipitation and increasing artificial exploitation affect the spring flux attenuation.
出处
《人民黄河》
CAS
北大核心
2012年第9期37-40,共4页
Yellow River
