基于时差相关分析与回归模型的用水总量预测

发布时间：2019-03-03 17:08

【摘要】：依据最严格水资源管理制度中水资源开发利用红线指标——用水总量的影响因素特征及其变化规律,对其提出综合运用时差相关分析与多元回归模型的预测分析方法,该方法特点是在时差相关分析法鉴别出先行影响因子的基础上,再建立多元回归模型预测用水总量。以太湖流域为实例,通过时差相关分析法筛选出与区域用水总量相关的人口、产业结构比、农田面积、人均用水量、工业用水量等8个预警指标,再借助SPSS软件建立多元线性回归分析预测模型,模型检验表明该方法具有较好的拟合度和预测精度。通过该模型方法预测了太湖流域2015年及2020年用水总量,并分析了其变化趋势,所用方法及其计算成果可为太湖流域水资源承载能力评估提供参考依据。
[Abstract]:According to the influencing factors and changing rules of total water consumption, the red line index of water resources development and utilization in the strictest water resources management system, this paper puts forward the prediction and analysis method of comprehensive utilization time difference correlation analysis and multiple regression model. The characteristic of this method is to establish a multiple regression model to predict the total amount of water on the basis of identifying the antecedent influencing factors by the method of time difference correlation analysis. Taking Taihu Lake basin as an example, eight early warning indexes, such as population, industrial structure ratio, farmland area, per capita water consumption and industrial water consumption, were selected by means of time difference correlation analysis, which are related to the total amount of water used in the region. With the help of SPSS software, the prediction model of multiple linear regression analysis is established. The model test shows that the method has good fitting degree and prediction precision. The total amount of water used in the Taihu Lake Basin in 2015 and 2020 is predicted by this model method, and its changing trend is analyzed. The method used and its calculation results can provide a reference basis for the assessment of the carrying capacity of water resources in the Taihu Lake Basin.
【作者单位】：扬州大学;
【基金】：国家自然科学基金项目(51379181)
【分类号】：TV213.4

【相似文献】