A prerequisite of a smooth operation of the ironmaking blast furnace is that the quality of the burden is stable. In blast furnaces where sinter is used as the (main) iron bearing material, its quality plays a crucial role in productivity and fuel economy. Simultaneously the corresponding factors must be considered for the sinter plant. The present paper studies the influence of three variables characterising the bedding piles and five sinter plant operation variables on sinter quality, sinter plant productivity, specific fuel consumption and share of cold return fines. Daily mean values for a period of five years of operation were used in the data driven modelling based on feedforward neural networks. The resulting models were found to describe the major changes in the outputs well. The input-output relations captured by the models were analysed by perturbing one input variable of the networks at a time and analysing the predicted behaviour of the outputs.
In developing data-driven models of complex real-world systems, a common problem is how to select relevant inputs from a large set of measurements. If the observations of the outputs to be predicted by the model are scarce, which may be the case if the outputs are indices determined in toilsome laboratory tests, strict constraints have to be imposed on the number of model parameters. In neural network modelling, this limitation in practice also restricts the number of hidden nodes as well as the number of input variables, since the dimension of the weight vector strongly depends on these. This paper presents a systematic method for data-driven modelling with feedforward layered neural networks, including a method for the selection of input variables. The method is illustrated on a problem from ironmaking industry, where sinter quality indices are predicted on the basis of raw material properties. Furthermore, an inversion technique of the resulting network models is proposed, where an optimization problem is solved to maximize the performance of the sintering operation by manipulating the inputs.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.