Evolutionary Data Driven Modeling and Multi Objective Optimization of Noisy Data Set in Blast Furnace Iron Making Process

Abstract

Data driven models are constructed for the tuyere cooling heat loss, total blast furnace gas flow, tuyere velocity, productivity, and coke rate for an operational blast furnace of an integrated steel plant by using evolutionary computation methods like bi objective genetic programming (BioGP) and evolutionary neural network (EvoNN), which serve as the objectives for their optimization. The models are used to compute the Pareto tradeoff between these conflicting objectives with the help of predator prey genetic algorithm well tasted for computing the Pareto optimality earlier. The results of optimization and the performances of these models are thoroughly analyzed and accessed and a discussion regarding these data driven models and their influence in the ironmaking process are presented. The results are also compared against a similar calculation performed with the commercial software KIMEME.

Collaborative Optimization Model of Blast Furnace Raw Materials and Operating Parameters Based on Intelligent Calculation

“…The normalized data is examined using box plots to check the distribution and identify possible outliers. Box plots can be used to display statistics such as median, quartiles, minimum and maximum values of the data, as well as outliers of the data 30) . In this study, box plots were used to analyze the data collected.…”

Advances in neural networks and potential for their application to steel metallurgy

“…This is much more streamlined than tuning the models manually but naturally leaves room to quantify even the best compromising model. This can be accomplished, for example, by using an information-based criterion [89,90]. This approach was also implemented later by Pettersson et al [91,92], and the final selected networks were shown, in each case, to perform better than an analytical model of the system.…”

“…This approach was also implemented later by Pettersson et al [91,92], and the final selected networks were shown, in each case, to perform better than an analytical model of the system. As noted by Mahanta and Chakraborti in their application of an EvoNN to blast-furnace operating data [89], there may be other constraints, for example, financial or practical, which can guide selection of a single optimal solution for a particular scenario.…”

“…They can fit data to a much wider array of functional types and have the benefit of only using mathematical functions from a predefined set, which allows the user to include some level of physical reasoning in the network, a notable inability with the rigid structure of NN methods. While GP methods do not seem to offer a substantial boost in accuracy or generalisation compared to appropriate NN methods [89,93,94], their benefit may come from the aforementioned flexibility and physical layer of intuition which can be bestowed upon them. This is particularly noticeable since most uses of basic NNs use them as simple arbitrary function approximators [4], but if partial analytical or empirical information is available then GP methods might serve to efficiently combine this prior knowledge with the learning characteristics favoured in NNs.…”

Genetic Algorithm‐Based Variable Selection in Prediction of Hot Metal Desulfurization Kinetics

“…Unfortunately, this is rarely available in the case of batch processes like hot metal desulfurization. Similar to Saxén et al, Mahanta et al applied an evolutionary neural network and bi‐objective genetic algorithm in the optimization of model structure for predicting several operational parameters in a blast furnace. They presented a pareto‐optimal set of input variables, i.e.…”