Research On Detection Technology Of PM2.5 Particle Size Distribution Using Total Scattering Method

At present,PM2.5 is the most important environmental problem in most cities of our country,which has a direct impact on human health and visibility.Therefore,study on the concentration and particle size distribution of PM2.5 has positive effects on the prevention and environmental treatment of respiratory diseases.Among the PM2.5 detection methods available,total light scattering method not only has the advantages of being used widly,high precision and convenient measurement,but also has the advantage of being non-contact.It has become the most widely adopted method in online monitoring of PM2.5 in recent years.Based on the principle of total light scattering method,the PM2.5 particle size distribution was studied from three aspects: the mathematical model,the calculation method and the experimental analysis.Main aspects of the research work in this paper are as follows:(1)Mie scattering theory,the Beer-Lambert extinction theory and their mathematical models are introduced.The extinction coefficient was solved and simulated through backward recursion method.The mathematical model of particle size distribution based on Beer-Lambert extinction theory and Mie scattering theory was derived.At last,the ill posed nature of the mathematical model is analyzed.(2)In the independent mode algorithm,the calculating speed of direct method is fast,but the accuracy is low,and the iterative method has high accuracy and good stability,but it is sensitive to initial value and noise.An improved algorithm based on direct method and iterative method is proposed,which makes full use of the advantages of two algorithms to make up for each other’s shortcomings,and improve the accuracy and efficiency of the algorithm;In the non-independent mode,the accuracy and stability of the characteristic parameters calculated by traditional objective function are poor,and the objective function is constructed by the Tikhonov smoothing function that essence is to obtain the best compromise between the accuracy and the stability of the function by adjusting the smoothness constraint.Under the new objective function,genetic algorithm(GA)and artificial fish swarm algorithm(AFSA)are used to inverse the particle size distribution.The experimental results show that in synthesis considerations on reconstruction accuracy and the reconstruction time,the objective function based on the Tikhonov function has good stability and reliability.(3)The Chahine iterative algorithm is used to retrieve,analyze and compare the distribution of the single peak R-R in the ultraviolet,UV-Vis,UV-Vis-NIR spectra,which determines the measurement range of the particle size of the typical spherical particle system in these three bands.(4)On the basis of theoretical analysis and numerical simulation,the modified Chahine iterative method,LMS iterative method,genetic algorithm and artificial fish swarm algorithm are verified by the measurement data of the national standard PM2.5 polystyrene particles.The experimental results show that the D50 relative error of GA and AFSA based on the Tikhonov smoothing function is less than the relative error of Chahine method and the LMS iteration method,which is more suitable for the fast and accurate measurement of the particle size.