Prediction of Gas Utilization Ratio Based on the Kernel Extreme Learning Machine

Abstract

Gas utilization ratio (GUR) is a significant indicator to measure the operation status and energy consumption of blast furnaces (BFs). Accurately predicting the GUR can reflect the actual operating status of the BFs and the consumption of the charge in real time. Kernel extreme learning machine (KELM) algorithm not only has the characteristics of fast computation speed of extreme learning machine (ELM), but also has better stability and generalization ability. This study applies KELM to investigate the relationship between GUR and some significant factors which affect GUR. An improved fruit fly optimization algorithm (IFOA) is used to optimize the parameters in the KELM model. The experimental results demonstrate that the prediction model based on the KELM has better prediction effect in forecasting accuracy and modeling time needed.