SYSTEM FOR PREDICTING RESIDUAL TAPPING AMOUNT OF BLAST FURNACE AND METHOD THEREFOR

摘要

A method for predicting a residual tapping amount of a blast furnace comprises: an operation data collection system for collecting a plurality of operation data; a learning system for generating and managing a neural network-based prediction model which outputs at least one prediction value associated with the residual tapping amount; and a prediction system for obtaining the at least one prediction value associated with the residual tapping amount of the blast furnace by using the plurality of operation data as input data for the prediction model, wherein the prediction model may include a plurality of neural network models for inputting one-dimensional, two-dimensional, or three-dimensional operation data, and an integrated layer for integrating features extracted through the plurality of neural network models.