Abstract
The paper presents optimization methods that are inspired by the mechanisms occurring in nature and their application in the determination of the values of signals controlling a copper flash smelting process. The furnace model that is used in the optimization process, was created on the base of artificial neural network. The control was aimed at the obtaining of the required values of SO2, CO2 and NOx concentrations in exhaust gases at a specific concentrate’s composition. The optimization process was based on a static furnace model and, therefore, the control consisted in the problem of static optimization. Block limitations were imposed on all the decision variables. The paper gives the theoretical background of each method, the way of implementation and the results obtained with its respective application. In addition, the results were compared with the results obtained by using one of deterministic methods.