基于比率方法和BP神经网络的柳州市工业需水量区间预测  

Interval Prediction of Industrial Water Demand in Liuzhou City Based on Ratio Method and BP Neural Network

作　　者：张志博 钟献词 谢仁奎[1,2,3] ZHANG Zhi-bo;ZHONG Xian-ci;XIE Ren-kui(College of Civil Engineering and Architecture,Guangxi University,Nanning 530004,China;Key Laboratory of Disaster Prevention and Structural Safety of Ministry of Education,Guangxi University,Nanning 530004,China;Guangxi Key Laboratory of Disaster Prevention and Mitigation and Engineering Safety,Guangxi University,Nanning 530004,China)

机构地区：[1]广西大学土木建筑工程学院,广西南宁530004 [2]广西大学工程防灾与结构安全教育部重点实验室,广西南宁530004 [3]广西大学广西防灾减灾与工程安全重点实验室,广西南宁530004

出　　处：《水电能源科学》2022年第4期54-57,共4页Water Resources and Power

基　　金：国家自然科学基金项目(11872155);广西高校高水平创新团队及卓越学者计划(202006)。

摘　　要：鉴于工业用水量测量的不确定性,其区间预测研究具有实际意义,基于比率方法和BP神经网络,构建了柳州市工业需水量区间预测模型。以2007~2013年数据为训练集,预测了2014~2019年柳州市工业需水量区间,引入综合指标对预测区间的适用性进行分析,发现置信水平80%情况下,预测模型的可靠性良好。进一步预测了2025年柳州市工业需水量,结果表明2025年柳州市工业需水量维持在3.378×10^(8)m^(3)左右,置信水平80%条件下,工业需水量波动区间为2.924×10^(8)~3.832×10^(8)m^(3)。该结果对柳州市水资源规划具有一定的理论和实践意义。In view of the uncertainty for the measurement of industrial water demand,the research on interval prediction has practical significance.Based on the ratio method and BP neural network,this paper established an interval prediction model of industrial water demand in Liuzhou City.Taking the data from 2007 to 2013 as the training set,the intervals of industrial water demand in Liuzhou from 2014 to 2019 were predicted.Comprehensive index was introduced to analyze the applicability of the prediction interval.It is found that the reliability of the prediction model is good when the confidence level is 80%.The industrial water demand of Liuzhou in 2025 is further predicted.The results show that the industrial water demand of Liuzhou in 2025 will be about 3.378×10^(8)m^(3).Under the condition of 80%confidence level,the range of industrial water demand is 2.924×10^(8)~3.832×10^(8)m^(3).The results have a certain theoretical and practical significance for Liuzhou water resources planning.

关 键 词：工业需水量 BP神经网络 综合指标 柳州市 区间预测 

分 类 号：TV213.9[水利工程—水文学及水资源]