基于炉内温度分布的NO_x排放特性的神经网络模型  

Neural Network Model on NO_x Emission with Furnace Temperature Distribution in a Boiler

作　　者：王淅芬[1] 罗自学[2] 周怀春[2] 

机构地区：[1]武汉科技大学机械自动化学院,湖北武汉430081 [2]华中科技大学煤燃烧国家重点实验室,湖北武汉430074

出　　处：《能源技术》2009年第3期133-136,148,共5页Energy Technology

基　　金：国家863计划项目资助(No.2007AA05Z306)

摘　　要：电站锅炉燃烧产物NOx的生成受煤种、锅炉负荷、配风方式、过剩空气系数、炉膛温度以及其分布的均匀性等多种因素的制约非常复杂。借助优化燃烧调整试验数据,将燃烧过程三维温度场可视化检测信息如炉膛断面最高温度作为重要原始信息,建立反映锅炉NOx排放特性的神经网络模型,并对此模型进行了校验。结果表明,该模型能准确预报锅炉在不同工况下的NOx排放特性,结合全局寻优技术,可为大型电厂锅炉低NOx排放提供稳态优化运行方案。The NOx generation mechanism is complicated, affecting by many factors, such as coal quality, boiler load, air distribution, excess air ratio, furnace temperature and temperature distribution uniformity, etc. Based on combustion adjustment experiment in-situ, combining the three-dimensional temperature distribution （such as furnace cross section highest temperature）, a artificial neural network model predicting the NOx emission was developed and verified. The modeling results show that such a model can predicate the NOx emission under various operating conditions, if combined with the optimization algorithm, the optimize operating parameters can be find to decrease the NOx emission.

关 键 词：锅炉 NOx 炉膛断面最高温度 神经网络 

分 类 号：TK229.6[动力工程及工程热物理—动力机械及工程]