摘要
选取制约河北省再生水利用的经济投资、设施建设和用水因素三方面9个指标,通过使用粒子群优化算法支持向量回归机PSO-SVR模型,建立了自变量与再生水利用量之间的非线性函数映射关系,并对河北省再生水利用量进行了预测分析。结果表明,PSO-SVM模型,具有较好的预测精度与泛化能力,优于PCR回归模型和逐步回归模型。运用该模型对2020年、2025年河北省再生水利用量进行了预测,并计算了再生水利用率,再生水利用量影响因素的敏感性分析表明,用水状况因素指标对再生水利用量的影响作用最大,是影响河北省再生水利用量的主要因素。
The development and utilization of reclaimed water is one of the important ways to alleviate the shortage of water resources.The establishment of a scientific and effective prediction model of reclaimed water utilization has important guiding significance for the development and utilization of reclaimed water.In this paper,9 indicators which restrict economic investment,facility construction,and water use factors that restrict the utilization of recycled water in Hebei Province are selected,and independent variables are established by using PSO-SVR model.The non-linear function mapping relationship with the reclaimed water utilization,and the prediction analysis of the reclaimed water utilization in Hebei Province.The results show that the PSO-SVM model have better prediction accuracy and generalization ability,better than PCR regression model and stepwise regression model.The model was used to predict the reclaimed water utilization in Hebei Province in 2020 and 2025,and calculate the reclaimed water utilization rate.The sensitivity analysis of the influencing factors of reclaimed water utilization shows that the indicators of water use status have the greatest effect on the reclaimed water utilization,and are the main factors affecting the reclaimed water utilization in Hebei Province.
作者
余鹏明
管孝艳
陈俊英
YU Pengming;GUAN Xiaoyan;CHEN Junying(Department of Irigation and Drainage,China Institute of Water Resources and Hydropower Research,Bejing 100048,China;Key Laboratory of Agricultural Soil and Water Engineering in Arid and Semiarid Areas,Ministry of Education,Northuest A&F Unirersity,Yangling,Shaanxi 712100,China)
出处
《水利与建筑工程学报》
2020年第4期70-75,共6页
Journal of Water Resources and Architectural Engineering
基金
国家重点研发计划课题(2017YFC0403503)
国家自然科学基金项目(51979234
51409221)。
关键词
再生水
PSO-SVR模型
潜力预测
敏感性分析
reclaimed water
PSO-SVR model
potential prediction
sensitivity analysis
Hebei Province
