摘要
针对阜新电厂200 MW机组燃煤锅炉进行了多工况热态测试,获得了飞灰含碳量的现场数据样本,运用Back Propagation(BP)神经网络和Levenberg-Marquardt(LM)算法建立了电站锅炉飞灰含碳量的软测量模型,并构造了飞灰含碳量的测量系统. 在此基础上开发了电站锅炉燃烧优化系统,实现了阜新电厂200 MW机组锅炉燃烧优化控制.
The unburned carbon content in the fly ash of a 200 MW unit coal burning boiler at Fuxin Power Plant is tested and data specimens have been gained. By taking back propagation (BP) neural network based on Levenberg-Marquardt (LM) algorithm, the soft measurement model of the unburned carbon content in fly ash is established. The measurement system of the unburned carbon content in the fly ash is made up of the soft measurement model. With the help of it, the optimization system of utility boiler combustion has been developed, which is used to realize optimal adjustment of a 200 MW unit coal burning boiler at Fuxin Power Plant.
出处
《过程工程学报》
EI
CAS
CSCD
北大核心
2004年第6期549-553,共5页
The Chinese Journal of Process Engineering
基金
辽宁省自然科学基金资助项目(编号:20022097)
关键词
锅炉
BP神经网络
飞灰含碳量
软测量
燃烧优化
boiler
BP neural network
unburned carbon
soft measurement
burning optimization