Optimization of Phosphorous in Steel Produced by Basic Oxygen Steel Making Process Using Multi-Objective Evolutionary and Genetic Algorithms

Abstract

Effective implication of evolutionary techniques in primary steel making process optimization is due, although the scope of its utilization in this sector is enormous. The multi-objective optimization technique is adopted in this paper for addressing important issue of achieving low phosphorus content along with high-end blow temperature in basic oxygen steel making process. The intend of this study is to predict optimal solution for the conflicting objectives namely minimization of end blow phosphorous content and maximization of end blow bath temperature considering initial hot metal temperature, hot metal tonnage, scrap, lime, and dolomite quantity as input variables. The meta-models have been created using evolutionary neural network (EvoNN) algorithm through training of a data set containing 120 steel plant data entries concerning different input as well as output variables. The analysis of Pareto front reveals that a low level of end blow phosphorous content is possible up to a certain critical value of end blow temperature. This study has a strong potential for providing an effective guideline for maintaining bath temperature just less than a critical temperature level during BOS steel making in order to control the phosphorous reversion along with attainment of necessary super heat.