Prediction for permeability index of blast furnace based on VMD–PSO–BP model

The permeability index is one of the important production indicators to monitor the operation of blast furnace. It is crucial to grasp the trends of changes in the new permeability index in time. For the complex vibration spectrum of the permeability index, a prediction model of the permeability index based on the VMD–PSO–BP (variational mode decomposition–particle swarm optimization–back propagation) method was proposed. Firstly, the key factors that affect the permeability index of blast furnace were studied from multiple perspectives. Then, the permeability index was divided into multiple sub-modes based on the difference of frequency bands by the VMD algorithm, and a PSO–BP prediction model was established for each sub-mode. Finally, the prediction results of each sub-mode were summed to obtain the final one. The results show that the composite prediction accuracy by using the VMD algorithm is 3% higher than that of the traditional prediction method, which has better applicability.