Burner Flame Characterization Based On Electrostatic Signal And Digital Image

At the critical low-carbon-energy transition stage,an in-depth study on the combustion characteristics of fuels is essential for the early realization of "carbon neutrality".The development of the commonly used detection methods and the electrical properties of gaseous fuel and biomass fuel flames were reviewed.Methane and biomass combustion experiments were carried out to study the flame shape and electrical characteristic parameters based on flame images and electrostatic signals.The main research contents are as follows:A method combining empirical mode decomposition(EMD)and wavelet threshold processing was proposed to denoise the electrostatic signals of flames measured by a non-intrusive electrostatic sensor array.The energy entropy of EMD of the electrostatic signals was used to characterize the motion complexity of charged particles in flames.The spatial distribution of the concentration of charged particles in a flame and the oscillation frequency and diffusion velocity of a flame were analyzed,respectively,in terms of the root mean square(RMS)value,the dominant frequency of the power spectral density(PSD)and the cross-correlation function of the electrostatic signals.Experimental results indicate that the fule flow condition and the intensity of chemical reaction greatly influence the motion complexity of charged particles in a flame.The highest concentration and motion complexity of charged particles are measured in the middle part of methane flams,while a relatively even spatial distribution of charged particles in biomass flame is determined.In order to obtain the burning area and shape of a flame accurately,an image segmentation method combining Canny edge detection algorithm and morphological processing was adopted to exact the length,width and normalized gray level of a flame under different combustion conditions by considering the contributing factors of these characteristic parameters.The oscillation frequency of the flames was also determined by spectral analysis of the cross-correlation coefficient of image sequences.The displacement of the maximum width of a flame was used to calculate the diffusion velocity of a flame.Experimental results suggest that the flame oscillation frequency increases with the fuel flow rate and the premixed flame diffusion velocity reaches the maximum value when the equivalence ratio is slightly greater than 1.