Soot propensity by image magnification and artificial intelligence

Abstract This paper presents the results of two novel approaches to measure the soot propensity of a flame and their comparison with the Line of Sight Attenuation (LOSA) method. Both approaches are based on the detection of the Smoke Point Height (SPH), concept used to determine when a flame is in a sooting state. The first approach is based on the detection of morphological changes in the flame, identified through their amplification via the Eulerian Video Magnification algorithm. Results show an effective amplification of the flame geometry, allowing the visualization of variations on the flame tip unable to be detected by the naked human eye and therefore the detection of SPH. The second approach is based on the application of Artificial Intelligence models to classify flame images regarding their sooting propensity, taking advantage of the knowledge acquired from a referential data set. Both approaches provide an accurate classification when compared to the conventional method of LOSA. Furthermore, both approaches show a greater implementation potential in practical combustion devices than the conventional method of LOSA, due to their reduced hardware and technical requirements.

[1]  Fengshan Liu,et al.  Experimental and numerical study of the effects of the oxygen index on the radiation characteristics of laminar coflow diffusion flames , 2013 .

[2]  Andrea Masiero,et al.  Advances on multivariate image analysis for product quality monitoring , 2013 .

[3]  Hongsheng Guo,et al.  Numerical study on the influence of hydrogen addition on soot formation in a laminar ethylene-air diffusion flame , 2006 .

[4]  C. McEnally,et al.  Sooting tendencies of nonvolatile aromatic hydrocarbons , 2009 .

[5]  Glassman,et al.  The Effect of Oxygen Concentration on Sooting Diffusion Flames , 1980 .

[6]  K. Cen,et al.  Study on the detection of three-dimensional soot temperature and volume fraction fields of a laminar flame by multispectral imaging system , 2016 .

[7]  Christopher M. Bishop,et al.  Neural networks for pattern recognition , 1995 .

[8]  Peter Norvig,et al.  Artificial Intelligence: A Modern Approach , 1995 .

[9]  Pal Toth,et al.  Image-based deep neural network prediction of the heat output of a step-grate biomass boiler , 2017 .

[10]  Hsueh-Chia Chang,et al.  Determination of the wavelength dependence of refractive indices of flame soot , 1990, Proceedings of the Royal Society of London. Series A: Mathematical and Physical Sciences.

[11]  Robert J. Santoro,et al.  Soot particle measurements in diffusion flames , 1983 .

[12]  Huaichun Zhou,et al.  Experiments investigation on 2D distribution of soot temperature and volume fraction by image processing of visible radiation , 2017 .

[13]  Javier Ballester,et al.  Diagnostic techniques for the monitoring and control of practical flames , 2010 .

[14]  Marcus Aldén,et al.  Combustion at the focus: laser diagnostics and control , 2005 .

[15]  Ronald K. Hanson,et al.  Soot pyrometry using modulated absorption/emission , 2001 .

[16]  Fengshan Liu,et al.  Effects of simultaneous hydrogen enrichment and carbon dioxide dilution of fuel on soot formation in an axisymmetric coflow laminar ethylene/air diffusion flame , 2016 .

[17]  John F. MacGregor,et al.  Multivariate image analysis in the process industries: A review , 2012 .

[18]  Chun Lou,et al.  Two-dimensional distributions of temperature and soot volume fraction inversed from visible flame images , 2013 .

[19]  Hugo O. Garces,et al.  Radiation measurement based on spectral emissions in industrial flames , 2016 .

[20]  Waldemar Wójcik,et al.  Advanced diagnostics of industrial pulverized coal burner using optical methods and artificial intelligence , 2012 .

[21]  Hua Wei Huang,et al.  Flame colour characterization in the visible and infrared spectrum using a digital camera and image processing , 2008 .

[22]  R. Kerr.  Global warming. Soot is warming the world even more than thought. , 2013, Science.

[23]  Patrick Haffner,et al.  Support vector machines for histogram-based image classification , 1999, IEEE Trans. Neural Networks.

[24]  Jean-Louis Consalvi,et al.  Unified behavior of soot production and radiative heat transfer in ethylene, propane and butane axisymmetric laminar diffusion flames at different oxygen indices , 2016 .

[25]  Guillaume Legros,et al.  Simultaneous soot temperature and volume fraction measurements in axis-symmetric flames by a two-dimensional modulated absorption/emission technique , 2015 .

[26]  Artur J. Ferreira,et al.  Characterization of Combustion Chemiluminescence: An Image Processing Approach☆ , 2014 .

[27]  Minxia Luo,et al.  A hybrid approach combining extreme learning machine and sparse representation for image classification , 2014, Eng. Appl. Artif. Intell..

[28]  Fengqi Si,et al.  On-line flame signal time series analysis for oil-fired burner optimization , 2015 .

[29]  Ping Zhang.  Model Selection Via Multifold Cross Validation , 1993 .