摘要
结垢是循环冷却水系统中常见的水质故障,人们常用水质判断指数来判断循环冷却水水质的结垢趋势。通过对某石化公司循环冷却水系统生产运行数据的分析,选取了对黏附速率影响较大的水质参数,借助神经网络的非线性映射、泛化及容错能力,基于BP神经网络建立了黏附速率的预测模型。利用该模型对循环冷却水系统一定周期黏附速率的预测结果较好,说明该方法可行,具有很好的应用前景。
Scaling is a common water quality fault in circulating cooling water systems.Water quality index is usually used for judging the scaling tendency of circulating cooling water quality. Based on the analysis on the production operation data of the circulating cooling water system in a petrochemical company, the water quality parameters, which have greater effect on scaling rate, have been selected for establishing the prediction model of scaling rate. It is based on BP neural network that has non-linear mapping, generalization, and fault tolerance capacities. This pre- diction model for predicting the scaling rate of the circulating cooling water system in a certain period has achieved good results. It shows that this method is feasible, having very good application prospect.
出处
《工业水处理》
CAS
CSCD
北大核心
2012年第2期44-46,共3页
Industrial Water Treatment
基金
天津市科技创新专项资金项目(05FZZDGX00300)
天津市教委滨海新区双百科技特派员科技专项(SB20080070)
关键词
循环冷却水
黏附速率
预测模型
神经网络
circulating cooling water
scaling rate
prediction model
neural network
