BP神经网络在供水管网余氯变化中的研究  

Study on the Changes of Residual Chlorine in Water Supply Pipeline Based on BP Neural Network

作　　者：赵加斌 ZHAO Jiabin(Shanghai Municipal Engineering Design and Research Institute(Group)Co.,Ltd.,Shanghai 200092,China)

机构地区：[1]上海市政工程设计研究总院(集团)有限公司,上海200092

出　　处：《佳木斯大学学报（自然科学版）》2023年第5期154-158,共5页Journal of Jiamusi University：Natural Science Edition

摘　　要：以机器学习为手段,用基础数据建立非机理性的BP神经网络模型。经实例验证,BP神经网络余氯水质模型能较好的模拟供水管网中余氯的变化,特别是对于南部供水管网,模型的检验阶段相关系数达到0.92,平均相对误差为17.20%,均方根误差约是0.13。以期为高品质供水的管网安全输送与水质保障研究提供方法借鉴和参考。This paper establishes the non-mechanistic model of BP(back-propagation)neural network.With the case study,the residual chlorine water quality model of BP neural network could get a good simulation effect on the changes of residual chlorine in the water supply pipeline.Especially in the stage of testing the model of the southern water supply pipeline,the correlation coefficient could be 0.92,and the average relative error is 17.20%,the average rms error is 0.13.This paper could provide some references for the study on the safety transmission of the water supply pipeline and water quality control of high-quality water.

关 键 词：BP神经网络 水质模型 供水管网 余氯 

分 类 号：TU991[建筑科学—市政工程] TP183[自动化与计算机技术—控制理论与控制工程]