Measurement of Equivalence Ratio Using Optical Flame Chemiluminescence Sensor of Turbulent Diffusion Flame

Transcription

1 Measurement of Equivalence Ratio Using Optical Flame Chemiluminescence Sensor of Turbulent Diffusion Flame Sewon Kim, Changyeop Lee, and Minjun Kwon Korea Institute of Industrial Technology, Cheonan, Chungnam, Korea {swkim, cylee, Kojima [3], Ikeda et al. [4] found the linear relationship between equivalent ratio and flame chemiluminescence ratio on laminar premixed flame, and they derived that this relationship can be applied to turbulent premixed flame and diffusion flame. Muruganandam et al. [5] derived the linear relationship between equivalent ratio and radical of specific chemical species, and they suggested the possibility of application to the combustion control of gas turbine combustor. The optical measurements are conducted at turbulent flames. The flame optical measurement is performed using optic fiber and photomultiplier tube. An extensive experimental works are conducted to derive the relationship between optic signal and flame operating condition, such as equivalence ratio. Abstract The objective of this study is to examine the relationship between the flame chemiluminescence and an equivalent ratio. In this experiment, flame optical signal in a furnace is measured using optic fiber and photomultiplier tube. The combustion system is composed of turbulent diffusion flame burner and furnace, and flame chemiluminescence is measured at various experimental conditions. In this study, the flame chemiluminescence of turbulent diffusion flame is measured at different measurement positions. The relationship between equivalent ratio and chemiluminescence intensity is experimentally investigated in the test furnace. The results shows that useful relationship between chemiluminescence intensity and equivalence ratio exists, and this leads to the successful application of this system for instantaneous measurement of equivalence ratio of the combustion system. Index Terms flame chemiluminescence, equivalence ratio, combustion control I. photo diode, II. FLAME CHEMILUMINESCENCE THEORY The analysis of light emitted by flames is the basis of many diagnostic techniques, and this technique is applied to determine equivalence ratio in this study. Fig. 1 is flame spectral analysis at UV and VIS region (wavelength: 200nm~550nm). There are three large peaks OH, CH, C2 respectively. The center wavelength of each peak are 308nm (OH), 431nm (CH), 515nm (C2) approximately. [6] In a recent study, the mainly targets of the radicals are OH, CH, C2. In particular, the radicals of CH and C2 have a relatively large peak. And the CO2 radical and blackbody radiation of soot are present in all wavelengths. INTRODUCTION This study is to develop monitoring and controlling techniques in order to increase the efficiency and reduce the NOx emissions of combustion system by measuring flame equivalence ratio in real time. [1] Thus, there are growing interests for precise and real time combustion control technologies. In particular, many researches have been conducted on the development of real time equivalence ratio sensor in order to increase the efficiency of combustion system such as industrial furnace, industrial boiler and domestic boiler. A responsive sensor is required for real time combustion control of combustion systems. However, the existing sensors, such as zirconia type oxygen sensor, cannot be applied to real time combustion control because they have flue gas flow time delay and time delay of sensor itself. Optical measurement technique of flame chemiluminescence is useful to apply to combustion control because of their short response time. Higgins et al. [2] derived the relationship between two radicals, OH, CH, and equivalent ratio. Thus they expressed the relationship as a function of temperature and pressure Laminar premixed flame spectroscopic analysis metal fiber burner, LNG, 32000kcal/h, Intensity CH OH C Wavelength [nm] Figure 1. Flame spectroscopic analysis. Manuscript received June 4, 2015, revised August 25, doi: /ijoee

2 data of each species is acquired and saved by control system. The blackbody radiation influence of soot radical depends on the temperature and the wavelength. Fig. 2 shows the Plank s curve for each temperature. It shows that the effect of temperature is significant above 400nm. Therefore when measuring the flame chemiluminescence, the blackbody radiation affects as a noise at long wavelength region. Main chemical species in combustion reaction are OH, CH, C2. The results of spectroscopic analysis are shown in Table I. Flow meter Fuel Flow meter 1000 Gas Analyzer Valve control signal 4~20mA Flow Signal 4~20mA DAQ & Control System (PC) Optical Sensor Voltage signal RS232 Figure 3. Schematic of experimental system. The optical sensor is comprised of optical fiber mounted in optical head, band pass filter, photomultiplier tube (), CCD camera, and spectrometer. The detailed schematic diagram of spectrometer system and radical measurement system is shown in Fig. 4. In addition, specification of three optical filters for three radical species, OH, CH, and C2 are given in Table II Regulator Flow meter LPG 1300K 1400K 1500K 1600K 1700K 1800K 1900K 2000K 1500 Intensity Air Regulator 2000 Furnace Burner Light Oil Collimator Wavelength [nm] Figure 2. Plank s curve for each temperature. Optical Head TABLE I. FORMATION ROUTES OF EXCITED RADICALS AND CHARACTERISTIC WAVELENGTHS Radical Reactions R1: OH R2: CH + O2 CO + OH Fiber Spectrometer Wavelength [nm] Bi convex lens 282.9, From flame H + O + M OH + M C2 signal CH signal OH signal R3: OH + OH + H OH +H2O Optical Head R4: C2H + O2 CO2 + CH R5: C2H + O CO + CH C2 R6: CH2 + C C2 + H2 513, Figure 4. Schematic of measurement system. CO2 R7: CO + O + M CO2 + M Continuous spectrum TABLE II. SPECIFICATION OF OPTICAL FILTERS CH 387.1, Filter For this study, measurements are conducted on turbulent flames generated using air staged diffusion burner. Fig. 3 is the schematic diagram of experimental system. The experimental system is comprised of turbulent diffusion flame burner, experimental furnace, fuel supply system, optical sensor, stack gas analyzer and control system. The air staged diffusion flame burner with maximum heat load 40,000kcal/h is used to generate turbulent flame and applied fuel is natural gas and light oil. Experimental furnace is 450mm wide and has a length of 1400mm. The Stack gas analyzer is type of chemical cell. It measures CO, NOX and O2. And the concentration Band pass Filter Center Wavelength peak transmission 308.7nm 17.22% CH nm 51.22% nm 66.22% OH III. EXPERIMENTAL CONDITIONS One by Three Fiber C2 The experimental variables in this study are heat load and equivalence ratio. In particular, the equivalence ratio can be expressed in terms of oxygen concentration of flue gas. The positions of optical sensor are shown in Fig. 5. As shown in the figure, measurements are performed in three positions, AGV (Axial Global View), RPV (Radial Parallel View), and RGV (Radial Global View). Optic fibers mounted in optical head are installed at these three different positions, and measurements are performed at various equivalence ratios. 204

3 as equivalence ratio increases. It is believed that momentum of fuel and air mixture decreases as equivalence ratio increases, thus reaction zone moves downstream as equivalence ratio increases. Flue gas Burner AGV Optical Head Quartz windows RGV RPV Figure 5. Location & focus of optical head. Both liquid fuel (light oil) and gaseous fuel (natural gas) are used in this study. The detailed experimental conditions of this study are given in Table III. TABLE III. EXPERIMENTAL CONDITIONS Figure 7. Distribution of CH for different equivalence ratio at gas flame. Fuel LPG Light-oil Heat Input 40,000 kcal/h Equivalence Ratio Fuel Pressure 1000mmAq, 12kgf/cm2 Gas Analyzer Data Rate 1Hz System Data Rate 100Hz RPV Location 50mm, 100mm 150mm, 200mm (From Burner Exit) Figure 8. Distribution of C2 for different equivalence ratio at gas flame. IV. RESULT AND DISCUSSIONS Distribution of radicals, OH, CH, and C2, are measured using optical sensor and photomultiplier tube at three different positions at various equivalence ratios. Fig. 6 shows the pictures of flames at different equivalence ratios in turbulent diffusion flames. As shown in the figure, both blue flame and yellow flame exist and the red flame moves downstream as equivalence ratio increases Figure 9. Picture of oil flame for equivalence ratio Figure 6. Picture of gaseous flames at different equivalence ratios. Fig. 7 and Fig. 8 show the distribution of CH radicals and C2 radicals in gaseous flames respectively, for different equivalence ratios. These radical distributions are obtained using band pass filter and CCD camera. In low equivalence ratio ranges, CH and C2 radicals mainly exist in flame front, but these radicals moves downstream Figure 10. Distribution of CH for each equivalence ratio at oil flame. 205

4 Figure 11. Distribution of C2 for each equivalence ratio at gas flame. Fig. 9 shows the pictures of flames at different equivalence ratios in turbulent diffusion liquid fuel flames. As shown in the figure, mainly yellow flame exists in light oil flame. It can be seen that flame length increases as equivalence ratio increases, and turbulence intensity increases. Fig. 10 and Fig. 11 show the distribution of CH radicals and C2 radicals in liquid fuel flames respectively, for different equivalence ratios. These radical distributions are obtained using band pass filter and CCD camera, same as that of gaseous flame. In these oil flames, main reaction zone does not move as the equivalence ratio increases, but the radical intensity decreases. Fig. 12 shows the results of spectrum analysis of gaseous and liquid flames at different equivalence ratios at three different measurement positions. The radical and chemiluminescence intensity for different equivalence ratio agree well with the predicted radical distribution data. In case of RPV position, the measurement was made at 5 different axial positions, and the results of spectrum analysis are shown in Fig. 13. As shown in the figure, chemiluminescence intensity shows similar pattern as that of Plank s curve. Fig. 14 shows the OH, CH, and C2 intensity changes at different equivalence ratios and at different measurement positions. The measurement was made using bandpass filter and photomultiplier tube. It can be seen that radical intensity decreases as equivalence ratio increases, but the reverse is true for gaseous flames at AGV position. It is because the yellow flame increases in downstream as equivalence ratio increases, meaning that the soot blackbody radiation affect the optical intensity. Figure 12. Spectrum analysis at each position. Figure 13. Spectrum analysis at all of RPV (Radial Parallel View). 206

5 Figure 14. Radical intensity at each position. Trade, Industry & Energy, Republic of Korea. (No ) V. CONCLUSION An extensive experimental works are conducted to derive the relationship between chemiluminescence intensity and equivalence ratio. Optic sensor consisted of optic fiber and photomultiplier tube is devised and chemiluminescence intensity of OH, CH, and C2 radicals are measured at different positions. Experimental study in turbulent diffusion flame using both gaseous fuel and light oil fuel are conducted, and the results are as follows. 1) In case of gas flame, high CH and C2 radical distribution zone moves downstream as equivalence ratio increase. 2) In light oil flame, chemiluminescence intensity decreases as equivalence ratio increases, but high radical distribution zone does not move downstream as the equivalence ratio changes. 3) In gaseous flame the effect of soot blackbody radiation can be observed in AGV position, but this effect cannot be observed in RPV, and RGV positions. 4) In light oil flame, peak radical intensity cannot be observed in AGV position, but strong OH, CH, and C2 radical peak intensity can be observed in RPV, and RGV positions due to soot blackbody radiation. 5) In RPV and RGV positions, OH, CH, and C2 radical intensity decrease as equivalence ratio increase. REFERENCES [1] [2] [3] [4] [5] [6] Sewon Kim was born in Seoul, Korea. He is granted a Ph.D. degree in Lowa State University, U.S.A. in His major is combustion engineering of energy system. He is working as a principal researcher in KITECH (Korea Institute of Industrial Technology) and developing technologies on energy efficiency and pollutants. Dr. kim is working in various mechanical academic society such as KSME (Korea Society of Mechanical Engineering) and Combustion Institute. ACKNOWLEDGMENT This work was supported by the Energy Efficiency & Resources Core Technology Program of the Korea Institute of Energy Technology Evaluation and Planning (KETEP), granted financial resource from the Ministry of S. Kim, C. Y. Lee, and M. Kwon, Real-Time integrated control system to improve boiler efficiency, Research of Energy Efficiency & Resources, Annual report, B. Higgins, M. Q. McQuay, F. Lacas, J. C. Rolon, N. Darabiga, and S. Candel, Systematic measurements of OH chemiluminescence for fuel-lean, high-pressure, premixed, laminar flames, Fuel, vol. 80, pp , J. Kohima, Y. Ikeda, and T. Nakajima, Spatially resolved measurement of OH, CH, and C2 chemiluminescence in the reaction zone of laminar methane/air premixed flames, Proc. of the Combustion Institute, vol. 28, pp , Y. Ikeda, J. Kojima, T. Nakajima, F. Akamatsu, and M. Katsuki, Measurement of the local flamefront structure of turbulent premixed flames by local chemiluminescence, Proc. of the Combustion Institute, vol. 28, pp , T. M. Muruganandam, B. Kim, R. Olsen, M. Patel, B. Roming, and J. M. Seizman, Chemiluminescence based sensor for turbine engines, in Proc. 39th Aerospace Sciences Meeting and Exhibit, S. Kim, C. Y. Lee, and M. Kwon, An experimental study on the measurement of fuel to air ratio using flame chemiluminescence, World Academy of Science, Engineering and Technology Energy and Power Engineering, vol. 8, no. 2,

6 Changyeop Lee was born in Jeonju, Korea. He is granted a Ph.D. degree in KAIST (Korea Advanced Institute of Science and Technology) in His major is precise measurement in harsh environments. He is working as a principal researcher in KITECH (Korea Institute of Industrial Technology) and developing technologies on optical measurements. Dr. Lee is working in various mechanical academic society such as OSK (Optical Society of Korea) and KSME (Korea Society of Mechanical Engineering). Minjun Kwon was born in Seoul, Korea. He is granted a master degree in Hanyang University in His major is Combustion. He is working as a researcher in KITECH (Korea Institute of Industrial Technology) and developing technologies on industrial combustion facilities. 208