Evolutionary Computations Based on an Artificial Bee Colony for the Analysis of Interval Data in the Problem of Air Pollution by Nitrogen Dioxide

Abstract

Evolutionary computation methods based on an artificial bee colony algorithm as one of the directions of computational intelligence are considered. It is shown that these methods are effective for the problems of analysis of interval data when the model of pollution by harmful emissions of motor vehicle exhaust gases is bild. The proposed calculational schemes made it possible to solve the parametric identification problem of the interval model of the distribution of pollution of the squat layer of the atmosphere with nitrogen dioxide. This is a component in the exhaust gases of vehicles. Verification of the constructed model was carried out on the test data obtained for the same area, but for variable conditions of traffic, which confirmed the sufficient prognostic properties of the model.