基于炉内火焰图像的燃烧诊断方法研究

针对现今电厂采用的锅炉燃烧监控系统所采集的火焰动态图像,提出了一种利于辅助分析的炉内火焰燃烧诊断方法。该方法分析了火焰图像的特点,提出了分析火焰稳定性判别的三个特征量,根据现场图像信号作出数据提取,并对提取结果进行了数据分析。提出一种燃烧稳定性判别方法,该方法利用以上三个特征量作为BP神经网络输入参数,得到输出确定为火焰稳定性系数,然后用模糊判别给出准确的燃烧稳定性综合评估。此方法利用了动态图像的差分特性,动态地分析燃烧过程中火焰锋面变化的状况,为现场锅炉监控人员提供了一种燃烧状态监测方法,方便了现场运行人员及时快捷地对现场状况做出准确迅速的判断和操作。通过对现场图像的人工分析和此方法判别结果比对,证明此方法具有很强的辅助分析功能。

Research on Combustion Diagnosis Based on Flame Image

Based on dynamic flame image of furnace,the way to analyze the combustion stability was present.By analyzing the characteristic of the flame image,three character values of image was proposes.Three character values of image was based on the research of dynamic flame image of furnace,and dealt of mass data.A method judging pulverized coal combustion stability by BP Neural Network and fuzzy pattern recognition was presented by using the three character values of image.Characteristic of dynamic image is utilized,and the states of flame were judged.This method can realize the intelligent diagnosis for boiler combustion.The result is as same as operator's judgment.The article applied a method for auto monitoring system to guide the operator making proper operation in short time.