Abstract
Silicon content forecasting models have been requested by the operational team to anticipate necessary actions during the blast furnace operation when producing molten iron, to control the quality of the product and reduce costs. This paper proposed a new algorithm to perform the silicon content time series up to 8 h ahead, immediately after the molten iron chemical analysis is delivered by the laboratory. Due to the delay of the laboratory when delivering the silicon content measurement, the proposed algorithm considers a minimum useful forecasting horizon of 3 h ahead. In a first step, it decomposes the silicon content time series into different subseries using the Maximal Overlap Discrete Wavelet Packet Transform (MODWPT). Next, all subseries forecasts were determined through Nonlinear Autoregressive (NAR) networks, and finally, these forecasts were summed to furnish the long-term forecast of silicon content. Using data from a real industry, we showed that the prediction error was within an acceptable range according to the blast furnace technical team.
Highlights
To increase the accuracy of prediction models using neural networks, we developed a hybrid algorithm based on Maximal Overlap Discrete Wavelet Packet Transform (MODWPT) decomposition and ANN models to forecast the content of silicon in molten iron
We proposed a long-term forecasting model of silicon content, which performed the decomposition of the time series into additive components using MODWPT
We used Nonlinear Autoregressive (NAR) neural networks to model each decomposed signal, and the long-term silicon content forecast was performed by adding each subseries forecasting
Summary
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Over the past two decades, several papers have proposed neural network models with different inputs to predict and even control the silicon content of molten iron and other parameters related to the quality of the blast furnace process. The authors of [10] eliminated the exogenous inputs, by decomposing the silicon content time series into different subseries from Empirical Mode Decomposition (EMD) After, these authors added the subseries forecasts through NAR neural network models.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
