摘要
本研究采用BP神经网络对污泥脱水调理剂的投加比组合进行优化。为了更全面体现对调理效果的综合评价,本研究引入层次分析法(AHP),确定板框处理产能、板框单批次运行时间、污泥运输处置成本等指标的权重。处理汇总后形成统一无量纲的综合评价值,作为神经网络的评价指标。结果表明,当阳离子PAM投加比为污泥干基的2.2‰,聚合硫酸铁投加比为污泥干基的18.5%,过氧化氢投加比为污泥干基的6.5%时,综合评价值最高。通过本优化方法所拟合的数据相关系数R2值为0.9916,具有较高的拟合度,适用于水质净化厂污泥调理药剂投加比组合的优化。
The BP neural network is used to optimize the dosage ratio combination of sludge dewatering conditioning agents.In order to comprehensively reflect the comprehensive evaluation of the conditioning effect,the Analytic Hierarchy Process(AHP)is introduced to determine the weights of indicators such as plate and frame processing capacity,single batch operation time of plate and frame,and sludge transportation and disposal cost.After processing and summarizing,a unified and dimensionless comprehensive evaluation value is formed as the evaluation index of the neural network.The results showed that when the dosage ratio of cationic PAM was 2.2‰of sludge dry basis,the dosage ratio of polymeric iron sulfate was 18.5%of sludge dry basis,and the dosage ratio of hydrogen peroxide was 6.5%of sludge dry basis,the comprehensive evaluation value was the highest.The correlation coefficient R2 value of the data fitted by this optimization method is 0.9916,which has a high degree of fit and is suitable for optimizing the dosage ratio of sludge conditioning agents in water purification plants.
作者
覃大松
容威
曾梓健
钟敏
Qin Dasong;Rong Wei;Zeng Zijian;Zhong Min(Shenzhen Deepwater Ecological Environment Technology Co.,Ltd.,Shenzhen 518000,China)
出处
《皮革制作与环保科技》
2024年第20期124-126,共3页
Leather Manufacture and Environmental Technology
关键词
市政污泥
污泥调理药剂
层次分析法
BP神经网络
municipal sludge
sludge conditioning agents
analytic hierarchy process
BP neural network
