基于BP神经网络的城市自来水厂供水量预测研究

杨存满; 鞠佳伟; 袁芳; 李晓尚; 兰华春

引用本文:	杨存满,鞠佳伟,袁芳,等.基于BP神经网络的城市自来水厂供水量预测研究[J].灌溉排水学报,,():-.
	Yang cunman,Ju jiawei,Yuan fang,et al.基于BP神经网络的城市自来水厂供水量预测研究[J].灌溉排水学报,,():-.

【打印本页】【下载PDF全文】【查看/发表评论】【EndNote】【RefMan】【BibTex】

过刊浏览高级检索

本文已被：浏览 2147次下载 0次
分享到：微信更多字体:加大+\|默认\|缩小-
基于BP神经网络的城市自来水厂供水量预测研究
杨存满¹, 鞠佳伟¹, 袁芳², 李晓尚³, 兰华春^1,4
1.清华苏州环境创新研究院;2.中环保水务投资有限公司;3.湘潭中环水务有限公司;4.清华大学环境学院北京

摘要:

针对目前城市自来水厂供水量预测与水源、气象等影响因素关系研究较少的问题,通过分析水源水的水文信息、水质参数、气象条件等与日供水量之间的相关性,建立水厂日供水量模型。基于湖南省湘潭市某自来水厂自控采集数据分析处理后,采用BP神经网络模型对水厂日供水量进行了预测,预测结果与实际情况具有很好的一致性, 预测误差小, 基本满足平均绝对百分比误差(MEAP)<5%的情况,能满足供水系统调度的实际需要。可见, 本预测神经网络模型是合理的, 为城市自来水厂日供水量预测提供了一种简单可行的思路和方法。

关键词: 自来水厂;日供水量;BP神经网络;预测;平均绝对百分比误差

DOI：

分类号:TU991.2

基金项目:

Prediction model of water-supply of urban waterworks based on BP neural network

Yang cunman¹, Ju jiawei¹, Yuan fang², Li xiaoshang³, Lan huachun^1,4

1.Research institute for environmental innovation(Suzhou), Tsinghua;2.General water of China Co., Ltd;3.Xiangtan Zhonghuan Water Affairs CO., Ltd;4.School of Environment, Tsinghua University

Abstract:

There is little research on the relationship between water supply prediction and water source /meteorological factors. Therefore, the daily water supply model of waterworks was established by analyzing the correlation between the hydrological information, water quality parameters and meteorological conditions of water source and daily water supply. After a waterworks automatic acquisition data analysis processing, the BP neural network model was used to predict the daily water supply of water plant. The prediction results are in good agreement with the actual situation, and the prediction error was little, basically meeting MEAP <5%, which can meet the actual demand of water supply system scheduling. It can be seen that the prediction neural network model is reasonable and provides a simple, feasible idea and method for daily water supply prediction of urban waterworks.

Key words: urban waterworks; daily water supply; BP neural network; prediction; MEAP