Optimization Of Regulation Algorithm About Particle Size Distribution And Analysis Of Error

All ikinds of different sources and sizes of acrosol particles exist in atmosphere. They all have an important impact on our health and life. A wide range of micro physical parameters can be retricvedri from multi-wavelenenth lidar.But it puts forward higher requirements to inversion algorithm of size distribution.In this paper size distribution inversion algorithm based multi-wavelength lidar was proposed. The algoriithm is based on the well-known regularization technique and combined extinction cocflicicient and backward scattering coefficient of 355nm. 532nm and 1064nm.Then the optmization scheme is put forward for solving the problem of the inversion results instability. The scheme make contribution to averaging the results of multiple sets of inversion,and selecting the proper percentage of interval which satisfied the minimum relative error of micro-physical parameters. In order to obtain the microphysical parameters of aerosol, this paper presents a particle inversion algorithm based on multi-wavelength lidar. The simulation results show that the relative error of the effective radius and volume concentration is within 10% under the distribution of three typical aerosol size distribution, the absolute error of the complex refractive index is within ± 0.04, the imaginary part error is about ± 0.005, the inversion cffect is much better.On the basis of the above optimization algorithm, the influence of the optical parameters of ceach channel on the inversion resulls is studied. Firstly, inversion effects of the single-channel optical paramters error is analyzed. When each optical input is perturbed by proportion 200?20% and inversion is repeated, the error of the micro-physics parameter inversion increased. Meanwhile the error of the extinction coefficient of 355nm and 532nm have the significant influence on the inversion result, the backscattcring coefficient is the middle,and the 1064nm backscattering coefficient is the lowest. So optical parameters of 355nm and 532nm need to be measured accurately when computing the micro-physical parameters. Secondly, the influence of multi-channel error on the inversion results is analyzed on the basis of single channel. The results show that the relative error of the effective radius is about 35% and the volume concentration is within 30% when 10% error is introduced of different channels .These conclusions can direct the design of the lidar system and assess the approximate range of the relative error of the micro physical parameters to some extent.Finally, the typical weather combined with the regularization inversion algorithm using the existed multi-wavelength lidar system are studied. The experiment results show that the algorithm has good inversion effect on the problem of inversion about the particle size distribution and the micro-physics parameters under different weather conditions.