Application Research Based on GA-FWA in Prediction of Sintering Burning Through Point

Abstract

he sintering burn through point (BTP) is the most important parameter in the sintering process, which is usually used to evaluate the quality of sintered products. However, due to the sintering process is a complex physical and chemical reaction process, the specific sintering end point cannot be detected. In this paper, the fireworks algorithm based on genetic algorithm (GA-FWA) is applied to optimize the parameters in support vector machines(SVM). A new intelligent optimization algorithm is proposed to optimize the sintering terminal prediction model of support vector machines. The model uses MATLAB software to test the actual production data of a steel factory. Through a large number of experiments, the average relative error of the experimental results is 0.0778%, and the average absolute error is 0.0843. The accuracy of the experimental results is obviously higher than the other methods of predicting the sintering BTP. It can accurately predict the position of the sintering end point, greatly improve the quality and production of the sinter in the steel plant, and also prolong the service life of the sintering equipment and reduce the production cost.