摘要
在火电厂循环水处理工艺中,浓缩倍率难以有效在线实时测量。而单纯依靠化验室对水质采样分析来计算控制,不仅需要配置实验设备,而且手工操作,存在实验参数、条件不易控制等问题。文中使用软测量技术,通过建立易测量的水质核心参数与浓缩倍率之间的BP神经网络模型,来对浓缩倍率进行估计,提高了浓缩倍率计算的准确性和水质控制效果。
In the thermal power plant circulating cooling water process, the concentrate rate cannot realized on-line measurement in real-time effectively, and it merely relying on sampling and analyzing for the water quality in the laboratory , and than to compute and control it . In this way, not only need equip the experimental, but need manual operation, and there are some problems such as experiment parameters ,experimental conditions is not easy to control. The paper using the soft-sensing technology, through building the neural network model between the easy measurement water quality key parameter and the concentrate rate, to estimate the concentrate rate, raised the accuracy of the concentrate rate computation and the effect of the water quality control.
出处
《电子设计工程》
2014年第7期57-59,共3页
Electronic Design Engineering
基金
陕西榆林市科技计划项目(gygg200711)
关键词
软测量
浓缩倍率
神经网络
建模
soft-sensing
concentrate rate
neural network
modeling
