Flame Temperature Measurement And Combustion Diagnosis Of Furnace Based On Spectral Analysis And Image Processing

It is important to realize the effective monitoring of combustion flame of furnace,to ensure the normal operation of the combustion equipment,to improve the combustion efficiency and to reduce pollutant emissions.The core of the effective monitoring of the combustion flame is the precise measurement of the combustion flame temperature and its radiation characteristics.In this paper,the combustion diagnosis of a 1030 MW coal-fired power plant boiler was conducted by using spectrometer system and image detection system.Firstly,a spectral separation method based on polynomial fitting was applied and the accuracy of the separation method was verified by numerical simulation and the Goodness-of-fit(GFC)coefficient.The results show that this method can effectively achieve the purpose of spectral separation,and the size of the signal-to-noise ratio directly affects the accuracy of the separation method.Then,the continuous flame radiation spectrum after spectral separation was analyzed by using the gray-body-band judgment method based on the two-color method to seek the gray radiation band of coal-burning flame.The results show that the coal combustion flame is in the range of 600~900 nm,which is consistent with the gray characteristics,and the accuracy of using the least squares method to calculate temperature is proved.Since the center response band corresponding to R and G is located in the non-gray body radiation range,this paper proposed to modify the image method temperature measurement by spectral analysis and the influence of ?G/?R on the temperature measurement was discussed by numerical simulation.The results show that the value of ?G/?R is changed by 20%,and the deviation of the temperature calculation caused by the change of ?G/?R is about 8%.So it is necessary to modify the image method temperature measurement by spectral analysis.Then,the characteristics of the flame radiation image were analyzed in detail,which indicates that the brightness of the flame radiation image has a positive correlation with the IR/IG and the emissivity of flame,but has a negative correlation with the flame temperature.Besides,the brightness flame of near the burners may be caused by soot particles that are volatilized from pulverized coal.Finally,a detailed diagnostic was conducted on the coal-fired power station,indicating that the temperature in the bottom and top of the furnace is lower,while the intermediate burners exhibit an approximate isothermal distribution.At the same time,the measurement temperature of 10 measuring points has a small fluctuation for a period of time.In order to monitor the combustion flame of the incinerator,and guide the staff effectively to adjust the air volume,feed quantity to achieve combustion optimization control,a portable spectrometer system was used to detect the temperature and emissivity of a flame in a 35 t/h municipal solid waste(MSW)incinerator.Firstly,the accuracy of the proposed method for the flame spectrum separation of MSW was proved,and the continuous spectrum of MSW flame was analyzed by using the graybody-band judgment method based on two-color method.The results show that in the wavelength range of 650~900nm,the refuse flame shows the gray body characteristic distribution.And the results of gray-band judgment before and after the spectral separation show that the existence of the discontinuous spectrum will cause serious distortion of the temperature and emissivity distribution,indicating that the necessity of the flame spectral separation.The relationship between the radiation intensity of alkali metal,the calculated temperature and the emissivity was studied.The results show that the alkali metal sodium and the two radiation intensity of the Potassium have the same tend with time and the variation of the flame temperature with time is in good agreement with the change tendency of the alkali metal radiation intensity with time,but there is no certain functional correspondence.At the same time,the flame temperature and emissivity show the opposite trend with time.At last,the combustion diagnosis of the MSW was conducted and made a detailed analysis by comparing the diagnosis results with the corresponding time DCS data.The results show that the change of primary air volume with time is consistent with the change trend of flame measurement temperature with time.The trend of flame measurement temperature and boiler load is also basically the same,but there is a delay in the relationship between the two,and the boiler load is delayed by the change of flame temperature.The relationship between the temperature and the flue gas composition was studied.It was found that there is a positive correlation among the temperature of the flame and the oxygen content,NOx and SO2 in the flue gas.The change of the content of NOx and SO2 is not only related to the flame temperature,but also to the change of primary air volume.