Research on prediction model of iron ore powder sintering foundation characteristics based on FOA-Catboost algorithm

Abstract

Sintered ore is the most important incoming iron material for blast furnaces in China. The foundation characteristics of sintering are an important basis for the sintering process of iron ore powder and a key indicator for evaluating the quality of iron ore powder. The chemical composition of iron ore powder directly affects the foundation characteristics of sintering. Based on the chemical composition and sintering characteristics of iron ore powder, the article adopts the FOA-Catboost algorithm to establish a prediction model for the foundation characteristics of single iron ore powder and mixed iron ore powder. Among the foundation characteristics prediction indicators of a single type of iron ore powder, the accuracy of assimilation temperature prediction is 95.65 %, the accuracy of fluidity index prediction is 95.23 %, and the accuracy of bonding phase strength prediction is 97.46. Among the foundation characteristics prediction indicators of mixed iron ore powder, the accuracy of assimilation prediction is 95.82 %, the accuracy of fluidity index prediction is 91.66 %, and the accuracy of bonding phase strength prediction is 93.7 %. On the basis of predicting the foundation characteristics of a single type of iron ore powder, the model further predicts the foundation characteristics of mixed iron ore powder, which can reduce the workload of laboratory experiments on the foundation characteristics of iron ore powder and reduce the amount of experiments in the process of finding the optimal proportion of mixed ore. Based on the foundation characteristics model of mixed ore, determine the optimal ratio of sintered ore powder to improve the quality of sintered ore. The establishment of the model has a significant contribution to improving experimental and production efficiency, reducing costs, improving sinter quality, and reducing environmental impact.