Methodology based on genetic optimisation to develop overall parsimony models for predicting temperature settings on annealing furnace

Abstract

Developing better prediction models is crucial for the steelmaking industry to improve the continuous hot dip galvanising line (HDGL). This paper presents a genetic based methodology whereby a wrapper based scheme is optimised to generate overall parsimony models for predicting temperature set points in a continuous annealing furnace on an HDGL. This optimisation includes a dynamic penalty function to control model complexity and an early stopping criterion during the optimisation phase. The resulting models (multilayer perceptron neural networks) were trained using a database obtained from an HDGL operating in the north of Spain. The number of neurons in the unique hidden layer, the inputs selected and the training parameters were adjusted to achieve the lowest validation and mean testing errors. Finally, a comparative evaluation is reported to highlight our proposal's range of applicability, developing models with lower prediction errors, higher generalisation capacity and less complexity than a standard method.

GA-PARSIMONY: A GA-SVR approach with feature selection and parameter optimization to obtain parsimonious solutions for predicting temperature settings in a continuous annealing furnace

“…Chen et al [32] also use an evolutionary approach to simultaneously optimize complexity and weights of learning vector quantification networks for bankruptcy prediction. Sanz et al [8] reported a novel GA-based optimization to create better overall parsimonious ANNs for predicting setpoints in an annealing furnace of a steel galvanizing industrial plant.…”

“…Developing parsimonious models presents a promising approach [7][8][9][10]. According to the parsimony criterion, the model with the least complexity that performs with similar accuracy is always preferred.…”

The consequences of eliminating the soaking section of a commercial hot-dip galvanising line for HSLA production and a control strategy for improving product consistency

“…The lack of historical data with this new configuration prevented the design of the new thermal cycles using data-driven models [9], thus the migration was progressive: for each cycle, first it was made an adjustment using simulations tools and lab tests with the Rhesca ® galvanizing simulator, then every new cycle had to be checked during one or more scheduled trials in the line before to be approved for its industrial use.…”