Particle Size Distribution Measurement Based On Light Scattering Method

Light scattering particle size distribution measurement is one of the important research branches in the particle optical detection field. The measurement accuracy of the particle size distribution measurement will be affected due to the complexity and diversity of the particle distribution. Therefore it is difficult to meet the measurement requirements of industrial production environment, which will also affect the wide application of light scattering particle size distribution measurement system in industrial production.By studying a lot of experimental methods of particle size distribution online measurement, especially the light scattering method, The particle size distribution can be obtained according to the scattering light intensity. Based on light scattering theory, the author proposed to utilize the approximate performance of the Eikonal approximation method and the Eikonal approximation equation to optimize the Mie Theory in calculating particles extinction coefficient. The verification indicates that the error of the reconstructed model is lower than 7%, and the extinction coefficient matrix is improved more than 48 times. A method of characteristic wavelength selection in all optical spectrum is used to analyze the relationship between the correlation degree of each vector and the amount of reconstruction information in the light source system. The verification results of the experiment show that the measurement error of the particle size distribution was within 9%.The key parameters such as scattering light intensity, scattering coefficient and absorption coefficient are difficult to be measured or calculated. The author put forward that Monte Carlo algorithm was needed to decompose the incident light and set the softsensing model in terms of calculating the key parameters. Invert the particle size distribution by using the soft-sensing model, of which the particle size was 2.08?m, and thus the inversion particle size was 2.176?m, which verified the feasibility of the data analysis system.