Multi-parameter Optical Diagnosis Method For Morphology Characteristics Of Soot Particle And Aggregate

The formation and evolution characteristics of soot particles have significant academic value and wide application background in low-carbon economy,environmental pollution,climate change,human health,wireless communication,infrared remote sensing,target recognition and other fields.The morphology characteristic parameters of soot particles are important reference factors for the study of soot formation in combustion.The development of soot diagnosis technology and accurate measurement of soot particle morphology parameters are very important for the study of soot formation mechanism and inhibition theory.Because of its non-invasive characteristics,optical diagnosis method is favored by the majority of researchers because it can obtain important information such as soot particle size,volume fraction and composition without interfering with flame combustion.However,most of the optical diagnostic methods for soot can only realize the inversion of a single kind of physical parameters,and there is a lack of research on the joint inversion of multiple kinds of physical parameters at home and abroad.And the optical diagnostic inversion method is essentially ill conditioned inverse problem,so the measurement accuracy is sensitive to the uncertainty of the prior parameters.Therefore,it is of great theoretical value and far-reaching scientific significance to overcome the ill conditioned characteristics of the inverse problem and develop an efficient,accurate and stable multi parameter collaborative inversion technology for the morphology characteristics of soot aggregate particles system.Based on the light scattering signal of continuous laser and the laser-induced incandescence signal of pulse laser,this paper systematically studies and analyzes the multi parameter collaborative inversion method of the morphology characteristics for quasi-sphere soot particle system and soot aggregate particle system.The specific research contents include the following five aspects:For a start,based on Mie scattering theory model and improved quantum particle swarm optimization(IQPSO)algorithm,a new method for joint inversion of particle size distribution and optical constants of single soot particle system is established,which overcomes the dependence of traditional light scattering method on prior information of optical constants.By comparing the multi value characteristics of different scattering signal schemes,it is revealed that the strategy of fusing multiple scattering signals and coupling multi wavelength enhances the mathematical constraints of solving the inverse problem.Numerical results show that the joint inversion method can accurately retrieve the particle size distribution and optical constants under 10% Gaussian noise.By analyzing the relationship between the number of scattering angles and the inversion accuracy,the number of scattering angles needed for joint inversion is optimized.Secondly,based on Rayleigh-Debye-Gans theory,the relative intensity calculation method of angle scattering is developed for polydisperse soot aggregate paticle system.The Covariance Matrix Adaptation Evolution Strategy(CMA-ES)algorithm is introduced to propose a new joint inversion method for aggreate size distribution and fractal dimension of soot aggregates.It is proved theoretically that the new inversion method does not need prior information of any model parameters,and overcomes the dependence of traditional light scattering method on the prior information of model parameters.In the case of 10% Gaussian noise,the accurate inversion of aggregate size distribution and fractal dimension of aggregates is realized by coupling multi wavelength strategy.The minimum number of scattering angles is obtained by analyzing multi value characteristics.Compared with IQPSO algorithm,CMA-ES algorithm greatly improves the efficiency and accuracy of joint inversion.Thirdly,based on the simplified laser-induced incandescence(LII)model at low laser fluence,a new model parameter(fractal dependent thermal accommodation coefficient)is defined to couple the aggregate structure,aggregate size distribution and thermal accommodation coefficient,and a new joint inversion method of primary particle size distribution and fractal dependent thermal accommodation coefficient is proposed.Based on the dual-model inverse problem framework which can avoid inverse crime,the primary particle size distribution and fractal dependent thermal accommodation coefficient of soot aggregate at room temperature can be accurately retrieved from the low-noise dual wavelength spectral LII decay signal,which overcomes the dependence of traditional LII method on the prior information of aggregate structure,aggregate size distribution and thermal accommodation coefficient.Then,the uncertainty analysis of three two-dimensional inversion subproblems reveals the limitations of the newly-developed multi parameter collaborative inversion method based on LII: the linear relationship between the mean value of primary particle size distribution and the fractal dependent thermal accommodation coefficient leads to the large uncertainty of the inversion results.By additional taking sublimation and thermionic emission into account in the simplified LII model,the mathematical characteristics of the inverse problem are reconstructed.The improved collaborative inversion method can accurately and stably retrieve the primary particle size distribution and fractal dependent thermal accommodation coefficient from the LII signal,and the method is suitable for large temperature range from room temperature to flame temperature.Finally,based on the planar LII experimental system of n-heptane flame,the experimental verification of the improved multi parameter LII reconstruction method of soot morphology characteristics is carried out.High pass filtering technology is used to shield the interference of laser refraction from the original measurement data and extract the measurement information of the pixel position in the soot area,which greatly reduces the number of pixels to be retrieved.The parameter field of primary particle size distribution and the fractal dependent thermal accommodation coefficient field were reconstructed from the processed two wavelength planar LII decay signal.The comparison between the inversion results of the sampling point and the TEM analysis results of the soot samples shows that the trend of the inversion primary particle size distribution is similar to the true value,and the mean values of the two are basically consistent,while the retrieval primary particle size distribution is narrower.