水资源与水工程学报

文章摘要

HUANG Yaoa,b,c,HUANG Suilianga,c.基于BP网络的人工沉床系统水质净化效果研究Journal of Water Resources and Water Engineering[J].,2011,22(6):

基于BP网络的人工沉床系统水质净化效果研究

Research on the water purification effectiveness based on modularized aero-adjustable submerged plant bed system of BP neural network

DOI：10.11705/j.issn.1672-643X.2011.06.224

中文关键词: BP神经网络人工沉床系统水质净化富营养化拟合

英文关键词: BP neural network modularized aero-adjustable submerged plant bed water purification eutrophication fitting

基金项目:国家自然科学基金(51079068)资助; 天津市科技计划重点项目(09ZCGYSF00400、08FDZDSF03402)资助; 国家水体污染控制与治理重大专项(2008ZX07314-005-001、2009ZX07209-001)资助

Author Name	Affiliation
HUANG Yaoa,b,c,HUANG Suilianga,c	HUANG Yaoa,b,c,HUANG Suilianga,c(a.Key Laboratory of Environmental Remediation and Pollution Control,b.Key Laboratory of Pollution Process and Environmental Criteria of the Ministry of Education,c.Numerical Simulation Group for Water Environment,College of Environmental Science and Engineering,Nankai University,Tianjin 300457,China)

Hits: 2176

Download times: 2

中文摘要:

人工沉床技术是一种新兴的生物-生态水体原位修复技术,水质净化过程具有非线性与不确定性,传统方法建立的数学模型难以精确模拟。本文以天津市外环河人工沉床实验2007年10月至2008年9月的水质实测数据为基础,建立BP神经网络水质模型,并与多元非线性回归方法进行比较。结果表明:建立的BP人工神经网络模型对不同水质指标拟合的平均相对误差分别为:化学需氧量(CODCr)0.1%,总氮(TN)1%,总磷(TP)0.6%。与多元非线性回归的拟合结果相比,平均相对误差更小,BP神经网络模型更适用于人工沉床系统出水水质的模拟。

英文摘要:

Modularized Aero-adjustable Submerged Plant Bed(MASPB) was a new kind of biology-ecological in-situ water remediation technology.Owing to the uncertainty and nonlinearity of the water purification process,it is hard to accurately simulate water quality with mathematical model which is established by traditional ways.Based on the water quality data in MASPB testing area of Tianjin from the October 2007 to the September 2008,Back Propagation(BP) model was established to get water quality simulation,and the re...

View Full Text View/Add Comment Download reader