Optimization of the Temperature Regime of Phosphorite Pellet Roasting Using the Coevolutional Bee Colony Algorithm

The article examines the scientific and practical problem of increasing the energy and resource efficiency of chemical and technological processes to produce phosphorite pellets from waste apatite–nepheline ores. A promising way to solve this problem is to find the optimal temperature regime for heating the charge in a roasting conveyor machine. The specifics of the optimization problem posed (nonlinearity of the objective function, large dimension of the search space, high computational complexity) determine the advisability of using stochastic methods, particularly the bee colony algorithm. To determine the values of “free” parameters that significantly affect the convergence of the algorithm, it was proposed to use the idea of coevolution, which consists in parallel running of several canonical bee colony algorithms with different “settings” for each subpopulation. Control of the chemical–technological system based on the results obtained will ensure energy-resource-saving operating conditions for the roasting conveyor machine.