A Multi-objective Optimization Model for Alloy Addition in BOS Process Based on ESN and Modified MOPSO

Abstract

This paper proposed a multi-objective optimization model to calculate the optimum adding amount of alloy during the process of basic oxygen steelmaking (BOS). In this model, one objective is the total costs of the alloys, and another objective is the total error of element contents. In order to establish the second objective, an echo state network (ESN) is adopted to predict the element contents. A modified multi-objective particle swarm optimization algorithm which has a chaos random mutation operator with Gaussian function proportions, called GMOPSO, is proposed to solve the alloy addition multi-objective optimization problem. Simulation results on practical data of BOS show that the costs optimized are lower than the actual costs, and the error of the element contents meets the demand for the steel products.