Dust Particle Size Distribution Inversion Based on the Multi Population Genetic Algorithm

The aerosol number size distribution is the main parameter for characterizing aerosol optical properties and physical properties, it has a major influence on radiation forcing. With regard to some disadvantages in the traditional methods, a method based on the multi population genetic algorithm (MPGA) is proposed and employed to retrieve the aerosol size distribution of dust particles. The MPGA principles and design are presented in detail. The MPGA has better performance compared with conventional methods. In order to verify the feasibility of the inversion method, the measured aerosol optical thickness (AOT) data of dust particles taken by a sun photometer are used and a series of comparisons between the simple genetic algorithm (SGA) and MPGA are carried out. The results show that the MPGA presents better properties when compared with the SGA with smaller inversion errors, smaller population size and fewer generation numbers to retrieve the aerosol size distribution. The MPGA inversion method is analyzed using the background day, dust storm event and seasonal size distribution. The method proposed in this study has important applications and reference value for aerosol particle size distribution inversion