摘要
一种基于光学原理的燃烧火焰/温度场测量装置,用以获得实时的炉内燃烧信息,以便实施洁净煤燃烧技术。文中以可视化火焰检测系统对电站锅炉燃烧火焰和温度场进行监测的研究。通过测量,得到了数值化的火焰/温度场信息,对燃烧火焰的图像进行了分析,提取了不同单色波波长下的火焰图像的平均灰度、方差、熵、火焰丰度、能量、最高灰度等特征量,计算得到了温度分布。为了建立锅炉排放与火焰参数及燃烧温度的关系,利用最小二乘支持向量机原理,以火焰参数为主要判据,将得到的表征燃烧的特征量作为最小二乘支持向量机的输入,对NOx排放量进行了预估。结果表明,估计值与实测值具有一致性。
This paper presents a study on the measurement of temperature distribution inside a pulverized coal boiler in a power station by using a novel flame/combustion visualization system. The combustion information is obtained from a boiler by this system and the clean combusting technology of coal is carded out. Flame images and temperature distributions reconstructed by the algorithm, providing both digital and graphical data. The characteristic parameters are defined as the average and the maximum grey levels, the variance, the entropy, the abundance and the energy of the flames acquired at different optical wave lengths, and their values are extracted from the experimental data. A feature space are formed consisting of the above six characteristic parameters for least square-support vector machines（LS-SVM） analysis, based on which the NOx emission is predicted by correlating emission and the above parameters of the flame. The results show that the predicted values and the measured values are in good agreement.
出处
《中国电机工程学报》
EI
CSCD
北大核心
2006年第12期161-165,共5页
Proceedings of the CSEE
基金
国家863计划项目(2003AA529290)~~
关键词
双色法
图像处理
最小二乘支持向量机
火焰可视化
NOX排放
two-color method
imaging processing
least square-support vector machines（LS-SVM）
flame visualization
NOx emission
