城市再生水年利用量预测分析  

Prediction Analysis of Annual Utilization of Urban Reclaimed Water

作　　者：侯金甫 方红远[1] 李艳明 程倩倩 HOU Jinfu;FANG Hongyuan;LI Yanming;CHENG Qianqian(College of Hydraulic Science and Engineering,Yangzhou University,Yangzhou 225009,China)

机构地区：[1]扬州大学水利科学与工程学院,江苏扬州225009

出　　处：《净水技术》2023年第4期95-102,共8页Water Purification Technology

基　　金：苏州市水利水务科技项目(2021010)。

摘　　要：城市再生水利用量预测对于城市再生水资源的优化配置至关重要。以苏州市为例,对影响再生水年利用量的因素进行相关性分析,筛选得到7个与再生水年利用量相关性较高的影响因素,并采用灰色GM(1,1)模型对各因素进行预测,预测结果作为BP神经网络的输入变量,利用BP神经网络非线性映射能力强、可自学习等优点,建立模型预测苏州城市再生水年利用量。模型验证分析表明:灰色模型和BP神经网络模型组合输出的预测结果与实际值之间的误差绝对值均小于1%,预测精度等级较高。最后应用该组合预测方法预测了苏州市2021年、2022年以及2025年的城市再生水利用量,以期为城市再生水利用量评估以及再生水利用规划、合理配置提供参考依据。The prediction of urban reclaimed water consumption is crucial for the optimal allocation of urban reclaimed water resources.Taking Suzhou City as an example,the correlation analysis of conducted on the factors affecting the annual utilization of reclaimed water,and 7 factors with high correlation with the annual utilization of reclaimed water were selected.The grey GM(1,1)model was used to predict the factors.The prediction results were used as the input variables of BP neural network.The BP neural network had the advantages of strong nonlinear mapping ability and self-learning ability,so the model was established to predict the annual utilization of reclaimed water in Suzhou.Model validation analysis showed that:The absolute errors between the actual values and the predicted results of the combination output of the grey model and the BP neural network model were less than 1%,so the prediction accuracy was high.Finally,the combined prediction method was used to predict the urban reclaimed water consumption in Suzhou City in 2021,2022 and 2025,in order to provide reference for the evaluation of urban reclaimed water consumption and the planning and rational allocation of reclaimed water utilization.

关 键 词：城市再生水利用量 预测 相关性分析 灰色模型 BP 神经网络 

分 类 号：TV13[水利工程—水力学及河流动力学]