In this paper, we applied the Particle Swarm Optimization (PSO) to the problem of the atmospheric pollution source localization. Localization is performed based on the information about concentration of the released substance in the atmosphere, recorded by a network of detectors deployed in the area. PSO algorithm searches the five-dimensional space in order to find the dispersion model parameters. The parameters are: the contaminant source coordinates (X, Y), release strength (Q) and the atmospheric transport dispersion coefficients (cry, crz). In the reconstruction, we employed the fast-running Gaussian plume dispersion model. We presented a sample configuration of the PSO algorithm, which task is to indicate the most likely parameters of contamination source in the shortest time. The efficiency of the algorithm was tested on the basis of the generated synthetic data.

71-91