Estimation of Flue Gas Temperature by Image Processing and Machine Learning Methods

Abstract

This paper presents an experimental study on the flue gas temperature estimation in small-scale nut coal-fired boiler. The flue gas temperature must be within a certain range depending on the fuel type, otherwise it causes corrosion in the boiler. Within the scope of this study, features were obtained from flame image. The flue gas temperature was estimated with these features and the SVR model. The flame image was taken with a CCD camera. At the same time, the reference flue gas temperature was taken with the flue gas analyzer. The flame image and temperature are recorded on the same computer. Flame image is converted to gray scale image and features are obtained. The intensity distribution of the flame image was used when obtaining the features. Two types of distribution were used for this process. The first is the histogram of the flame image to obtain a location independent intensity distribution. The second is to obtain a spatial intensity distribution using row and column sums. The attributes obtained from the combinations of these two type features were performed with 6 kinds of SVR models. The best results were obtained for the cubic SVR model for the feature extraction method in which the attributes obtained from both distributions were used together. In the proposed model the flue temperature (T ° C) was estimated with R = 0.97 accuracy. The results show that there is a high correlation between the flue gas temperature and the flame image.

Keywords

• Flue gas temperature estimation
• flame image
• support vector regression

Görüntü İşleme ve Makine Öğrenmesi Yöntemleri ile Baca Gazı Sıcaklığının Tahmin Edilmesi

Öz

Bu makalede, küçük ölçekli fındık kömürü yakıtlı brülörde baca gazı sıcaklığı tahmini ile ilgili deneysel bir çalışma sunulmaktadır. Baca gazı sıcaklığı yakıt türüne göre belli bir aralıkta olması gerekir aksi durumda kazanda korozyona sebep olmaktadır. Bu çalışma kapsamında alev görüntüsünden öznitelikler elde edilmiştir. Bu öznitelikler ve DVR modeli ile baca gazı sıcaklığı tahmin edilmiştir. Alev görüntüsü CCD kamera ile alınmıştır. Aynı zamanda referans baca gazı sıcaklığı, baca gazı analizörü ile alınmıştır.  Alev görüntüsü ve sıcaklık değeri aynı bilgisayara kaydedilmiştir. Alev görüntüsü gri seviye görüntüsüne çevrilerek öznitelikler elde edilmiştir. Öznitelikler elde edilirken alev görüntüsünün yoğunluk dağılımı kullanılmıştır. Bu işlem için iki tip dağılım kullanılmıştır. Birincisi görüntünün histogramı alınarak konumdan bağımsız yoğunluk dağılımının elde edilmesidir. İkincisi satır ve sütun toplamlarını kullanarak uzamsal yoğunluk dağılımının elde edilmesidir. Bu iki özniteliğin kombinasyonlarından elde edilen öznitelikler 6 çeşit DVR modeli ile gerçekleştirilmiştir. En iyi sonuçlar, her iki dağılımdan elde edilen özniteliklerin birlikte kullanıldığı öznitelik çıkarma yöntemi için kübik DVR modeli ile elde edilmiştir. Önerilen modelde baca sıcaklığı (T °C) doğruluk ile tahmin edilmiştir. Elde edilen sonuçlar baca gazı sıcaklığı ile alev görüntüsü arasında yüksek oranda bir ilişki olduğunu göstermektedir.

Anahtar Kelimeler

• Baca gazı sıcaklığı tahmini
• alev görüntüsü
• destek vektör regresyon

References

1. Baek, W. B., Lee, S. J., Baeg, S. Y., & Cho, C. H. Flame image processing & analysis for optimal coal firing of thermal power plant. https://doi.org/10.1109/ISIE.2001.931596
2. Baek, W. B., Lee, S. J., Baeg, S. Y., & Cho, C. H. (2001). Flame image processing & analysis for optimal coal firing of thermal power plant. {ISIE 2001: IEEE International Symposium on Industrial Electronics Proceedeing, Vols I-III}, {928-931}. https://doi.org/10.1109/ISIE.2001.931596
3. Bilgin, A. (2007). Kazanlarda enerji Verimliliği ve Emisyonlar. In VIII. Ulusal Tesisat Mühendisliği Kongresi (pp. 33–40). İzmir: TMMOB Makina Mühendisleri Odası. Retrieved from http://www1.mmo.org.tr/resimler/dosya_ekler/c676cb12b9a742f_ek.pdf
4. Bonefacic, I., Blecich, P., Bonefačić, I., & Blecich, -P. (2015). Two-color temperature measurement method using BPW34 PIN photodiodes. Engineering Review (Vol. 35). Retrieved from https://www.researchgate.net/publication/282818960
5. Conti, J., Holtberg, P., Diefenderfer, J., LaRose, A., Turnure, J. T., & Westfall, L. (2016). International Energy Outlook 2016 With Projections to 2040. https://doi.org/10.2172/1296780
6. GoldsWorthy, P., Eyre, D. J., & On, E. (2013). Value-in-use (VIU) assessment for thermal and metallurgical coal. The Coal Handbook: Towards Cleaner Production (Vol. 2). https://doi.org/10.1533/9781782421177.3.455
7. Golgiyaz, S., Talu, M. F., & Onat, C. (2016). Estimation of Excess Air Coefficient for Automated Feed Coal Burners with Image-Based Gauss Model. In International Conferece on Artificial Intelligence and Data Processing, IDAP’16 (pp. 528–531). Malatya-Turkey.
8. González-Cencerrado, A., Gil, A., & Peña, B. (2013). Characterization of PF flames under different swirl conditions based on visualization systems. Fuel, 113, 798–809. https://doi.org/10.1016/j.fuel.2013.05.077

9. González-Cencerrado, A., Peña, B., & Gil, A. (2012). Coal flame characterization by means of digital image processing in a semi-industrial scale PF swirl burner. Applied Energy. https://doi.org/10.1016/j.apenergy.2012.01.059
10. Huang, B., Luo, Z., & Zhou, H. (2010). Optimization of combustion based on introducing radiant energy signal in pulverized coal-fired boiler. Fuel Processing Technology, 91(6), 660–668. https://doi.org/10.1016/j.fuproc.2010.01.015
11. Jiang, Z. W., Luo, Z. X., & Zhou, H. C. (2009). A simple measurement method of temperature and emissivity of coal-fired flames from visible radiation image and its application in a CFB boiler furnace. Fuel. https://doi.org/10.1016/j.fuel.2008.12.014
12. Krabicka, J., Lu, G., & Yan SMIEEE, Y. (n.d.). A Spectroscopic Imaging System for Flame Radical Profiling.
13. Li, B., Chen, F., & Shi, L. (2015). A coupling method of direct and inverse heat conduction problems for transient temperature calculation of a boiler drum. Dongli Gongcheng Xuebao/Journal of Chinese Society of Power Engineering, 35(2), 96–102. https://doi.org/10.1299/jtst.2016jtst0030
14. Li, N., Lu, G., Li, X., & Yan, Y. (2014). Prediction of pollutant emissions of biomass flames using digital imaging, contourlet transform and Radial Basis Function network techniques. Conference Record - IEEE Instrumentation and Measurement Technology Conference, 697–700. https://doi.org/10.1109/I2MTC.2014.6860832
15. Li, N., Lu, G., Li, X., & Yan, Y. (2016). Prediction of NOx Emissions from a Biomass Fired Combustion Process Based on Flame Radical Imaging and Deep Learning Techniques. Combustion Science and Technology, 188(2), 233–246. https://doi.org/10.1080/00102202.2015.1102905
16. Li, T., Zhang, C., Yuan, Y., Shuai, Y., & Tan, H. (2018). Flame temperature estimation from light field image processing. Applied Optics, 57(25), 7259. https://doi.org/10.1364/ao.57.007259
17. Liu, Z., Zheng, S., Luo, Z., & Zhou, H. (2016). A new method for constructing radiative energy signal in a coal-fired boiler. Applied Thermal Engineering, 101, 446–454. https://doi.org/10.1016/j.applthermaleng.2016.01.034
18. Lou, C., Zhou, H. C., Yu, P. F., & Jiang, Z. W. (2007). Measurements of the flame emissivity and radiative properties of particulate medium in pulverized-coal-fired boiler furnaces by image processing of visible radiation. Proceedings of the Combustion Institute, 31 II, 2771–2778. https://doi.org/10.1016/j.proci.2006.07.178
19. Mu, H., Li, Z., Han, Z., Li, J., Schlaberg, H. I., Liu, S., & Liu, S. (2015). Visualization Measurement of the Flame Temperature in a Power Station Using the Colorimetric Method. In Physics Procedia. https://doi.org/10.1016/j.egypro.2015.02.075
20. Parveen, N., Zaidi, S., & Danish, M. (2017). Development of SVR-based model and comparative analysis with MLR and ANN models for predicting the sorption capacity of Cr(VI). Process Safety and Environmental Protection, 107, 428–437. https://doi.org/10.1016/j.psep.2017.03.007
21. Peña, B., Bartolomé, C., & Gil, A. (2017). Analysis of thermal resistance evolution of ash deposits during co-firing of coal with biomass and coal mine waste residues. Fuel, 194, 357–367. https://doi.org/10.1016/j.fuel.2017.01.031
22. Qi, M., Luo, H., Wei, P., & Fu, Z. (2018). Estimation of low calorific value of blended coals based on support vector regression and sensitivity analysis in coal-fired power plants. Fuel, 236, 1400–1407. https://doi.org/10.1016/j.fuel.2018.09.117
23. Quej, V. H., Almorox, J., Arnaldo, J. A., & Saito, L. (2017). ANFIS, SVM and ANN soft-computing techniques to estimate daily global solar radiation in a warm sub-humid environment. Journal of Atmospheric and Solar-Terrestrial Physics, 155, 62–70. https://doi.org/10.1016/j.jastp.2017.02.002
24. TALU, M. F., ONAT, C., & DASKIN, M. (2017). Prediction of Excess Air Factor in Automatic Feed Coal Burners by Processing of Flame Images. Chinese Journal of Mechanical Engineering, 30(3), 722–731. https://doi.org/10.1007/s10033-017-0095-3
25. Vapnik, V. N. (2000). The nature of statistical learning theory. Springer. https://doi.org/10.1007/978-1-4757-3264-1
26. Wang, F., Wang, X. J., Ma, Z. Y., Yan, J. H., Chi, Y., Wei, C. Y., … Cen, K. F. (2002). The research on the estimation for the NOxemissive concentration of the pulverized coal boiler by the flame image processing technique. Fuel, 81(16), 2113–2120. https://doi.org/10.1016/S0016-2361(02)00145-X
27. Wang, Z., Liu, M., Dong, M., & Wu, L. (2017). Riemannian Alternative Matrix Completion for Image-Based Flame Recognition. IEEE Transactions on Circuits and Systems for Video Technology, 27(11), 2490–2503. https://doi.org/10.1109/TCSVT.2016.2587378
28. Wu, F., Zhou, H., Ren, T., Zheng, L., & Cen, K. (2009). Combining support vector regression and cellular genetic algorithm for multi-objective optimization of coal-fired utility boilers. Fuel, 88, 1864–1870. https://doi.org/10.1016/j.fuel.2009.04.023
29. Xiangyu, Z., Shu, Z., Huaichun, Z., Bo, Z., Huajian, W., & Hongjie, X. (2016). Simultaneously reconstruction of inhomogeneous temperature and radiative properties by radiation image processing. International Journal of Thermal Sciences, 107, 121–130. https://doi.org/10.1016/j.ijthermalsci.2016.04.003
30. Xiangyu, Z., Xu, L., yu, Y., Bo, Z., & Hongjie, X. (2018). Temperature measurement of coal fired flame in the cement kiln by raw image processing. Measurement: Journal of the International Measurement Confederation. https://doi.org/10.1016/j.measurement.2018.07.063
31. Zhou, H., Han, C. (1994). An exploratory investigation of the computer-based control of utility coal-fired boiler furnace combustion. In J. Eng. Therm. Energy Power (pp. 111–116).
32. Zhou, H., Tang, Q., Yang, L., Yan, Y., Lu, G., & Cen, K. (2014). Support vector machine based online coal identification through advanced flame monitoring. Fuel. https://doi.org/10.1016/j.fuel.2013.10.041