Recognition of the atmospheric contamination source localization with the Genetic Algorithm

Abstract

We have applied the Genetic Algorithm (GA) to the problem of the atmospheric contaminant source localization. The algorithm input data are concentrations of given substance registered by sensor network. To achieve rapid-response event reconstruction,the fast-running Gaussian plume dispersion model is adopted as the forward model. The proposed GA scans 5-dimensional parametersspace searching for the contaminant source coordinates (x,y), release strength (Q) and the atmospheric transport dispersion coefficients. Based on the synthetic experiment data the GA parameters likepopulation size, number of generations and the genetic operators best suitable for the algorithm performance are identified. We demonstrate that proposed GA configuration can successfully point out the parameters of abrupt contamination source. Results indicate the probability of a source to occur at a particular location with a particular release rate. The shapes of the probability distribution function of searched parameters values reflect the uncertainty in observed data.