Investigation Of Particle Size Measurement Based On MIE Scattering Theory

This paper combined with the optical scattering method and the inversion algorithm of particle size distribution measurement is studied. Research work of this paper is mainly about solid particle measurement technology research, and to improve the optical measurement platform, using MATLAB software, for the first time some of the data processing method and the inversion algorithm used in the study of particle measurement. The main content of the thesis are as follows:(a) On the basis of mastering Mie scattering theory, light scattering modules for single particles and particle systems are established, theoretical derivation is carried out, and a program is written to perform simulation verification with MATLAB as the platform,including simulation of Mie scattered light intensity and distribution law, simulation of angular distribution law of scattered light of particles in different sizes, and so on. The specification parameters of corresponding annular photodetectors are designed according to the particle size measurement ranges of 0.1μm-100μm, and simulation detection is performed to the scattered light intensity of their particles systems.(b) Typical constraint inversion algorithm and the latest non constraint inversion algorithm are researched. Constraint algorithm including R-R inversion algorithm, the constraint inversion algorithm including iteration Chahine inversion algorithm, iterative Tikhonov regularization inversion algorithm and TSVD non-negative inversion algorithm.The respective characteristics of the above algorithm are summarized. It is found out by simulation researches that the R-R inversion algorithm is easy to calculate, but it is necessary to suppose that the particle size distribution meets a certain function, which greatly limits its applications. Chahine inversion algorithm is not suitable for the calculation of particle size under 10 microns particles. Iterative Tikhonov regularization inversion algorithm and TSVDnon-negative inversion algorithm can obtain any particle swarm size distribution in theory,but is sensitive to noise, so that it easily leads to distortion of particle size measurement results.(c) The luminous energy distrbution at the time of 0.5% and 5% noise simulation measurement is introduces into the luminous energy column vector. The iterative Tikhonov regularization inversion algorithm and TSVD non-negative inversion algorithm mode is sensitive to noise, so that it easily to distortion of particle size measurement results. In depth research is conducted on this problem, the complexity of luminous energy distribution and the nonoperability of denoising the luminous energy in practical measurement are discussed, and five point three data smoothing algorithm is introduced to make smoothing processing to the inversion results. This effectively eliminates the influence if errors, and improves the precision and reliability of particle size measurement.(d) Aimed at the shortcoming of traditional Fourier measuring system, an improved inverse Fourier laser measurement system, the particle size of 4.9 mu mu m and 10.9 m standard polystyrene sample of a large number of measurement, and combined with particle size of iterative Tikhonov regularization inversion algorithm, to verify the measurement system is proposed in this paper and the feasibility and effectiveness of inversion algorithm.With the researches of this thesis, system theoretical analysis and experiment researches on the laser particle size test based on Mie theory are completed. This proves that the theoretical model of it is feasible, and corresponding experimental researches and results are reliable. It also solves the problem that influence on practical application of particle size test inverse algorithm from noise leads to measurement distortion, and satisfies the requirements on high precision, reliability and stability of particle size measurement in modern power industry. Experiments show that after smoothing processing, the error distribution reaches unanimity, and the errors will be further controlled through verification and calibration.