Cold rolling mill thickness control using the cascade-correlation neural network

The improvements in thickness accuracy of a steel strip produced by a tandem cold-roIling mill are of substantial interest to the steel industry. In this paper, we designed a direct model-reference adaptive control (MRAC)  scheme that exploits the natural level of excitation existing in the closed-loop with a dynamically constructed cascade-correlation neural network (CCNN) as a controller for cold roIling mill thickness control. Simulation results show that the combination of a such a direct MRAC scheme and the dynamically constructed CCNN significantly improves the thickness accuracy in the presence of disturbances and noise in comparison with to the conventional PID controllers.

Predicting the rolling force in hot steel rolling mill using an ensemble model

Accurate prediction of the roll separating force is critical to assuring the quality of the final product in steel manufacturing. This paper presents an ensemble model that addresses these concerns. A stacked generalisation approach to ensemble modeling is used with two sets of the ensemble model members, the first set being learnt from the current input-output data of the hot rolling finishing mill, while another uses the available information on the previous coil in addition to the current information. Both sets of ensemble members include linear regression, multilayer perceptron, and k-nearest neighbor algorithms. A competitive selection model (multilayer perceptron) is then used to select the output from one of the ensemble members to be the final output of the ensemble model. The ensemble model created by such a stacked generalization is able to achieve extremely high accuracy in predicting the roll separation force with the average relative accuracy being within 1% of the actual measured roll force.

Modeling and control of flatness in cold rolling mill using fuzzy petri nets

Today, having a good flatness control in steel industry is essential to ensure an overall product quality, productivity and successful processing. Flatness error, given as difference between measured strip flatness and target curve, can be minimized by modifying roll gap with various control functions. In most practical systems, knowing the definition of the model in order to have an acceptable control is essential. In this paper, a fuzzy Petri net method for modeling and control of flatness in cold rolling mill is developed. The method combines the concepts of Petri net and fuzzy control theories. It focuses on the fuzzy decision making problems of the fuzzy rule tree structures. The method is able to detect and recover possible errors that can occur in the fuzzy rule of the knowledge-based system. The method is implemented and simulated. The results show that its error is less than that of a PI conventional controller.

Effect of microstructure on the stability of retained austenite in transformation-induced-plasticity steels

Two Fe-0.2C-1.55Mn-1.5Si (in wt pet) steels, with and without the addition of 0.039Nb (in wt pet), were studied using laboratory rolling-mill simulations of controlled thermomechanical processing. The microstructures of all samples were characterized by optical metallography, X-ray diffraction (XRD), and transmission electron microscopy (TEM). The microstructural behavior of phases under applied strain was studied using a heat-tinting technique. Despite the similarity in the microstructures of the two steels (equal amounts of polygonal ferrite, carbide-free bainite, and retained austenite), the mechanical properties were different. The mechanical properties of these transformation-induced-plasticity (TRIP) steels depended not only on the individual behavior of all these phases, but also on the interaction between the phases during deformation. The polygonal ferrite and bainite of the C-Mn-Si steel contributed to the elongation more than these phases in the C-Mn-Si-Nb-steel. The stability of retained austenite depends on its location within the microstructure, the morphology of the bainite, and its interaction with other phases during straining. Granular bainite was the bainite morphology that provided the optimum stability of the retained austenite.

The effect of Nb, Mo, and A1 additions on the transformation behaviour and mechanical properties of thermomechanically processed C-Si-Mn trip steel

The effect of additions of Nb, A1 and Mo to Fe-C-Mn-Si TRIP steels on the final microstructure and mechanical properties after simulated thermomechanical processing (TMP) has been studied. Laboratory simulations of continuous cooling during TMP were performed using a quench deformation dilatometer, while laboratory simulations of discontinuous cooling during TMP were performed using a hot rolling mill. From this a comprehensive understanding of the structural and kinetic aspects of the bainite transformation in these types of TRIP steels has been developed. All samples were characterised using optical microscopy and XRD. The relationships between the morphology of bainitic structure, volume fraction, stability of RA and mechanical properties were investigated.

Effect of alloying elements on the microstructure property relationship in thermomechanically processed C-Mn-Si TRIP steels

The effect of additions of Nb, Al and Mo to Fe-C-Mn-Si TRIP steel on the final microstructure and mechanical properties after simulated  thermomechanical processing (TMP) has been studied. The laboratory simulations of discontinuous cooling during TMP were performed using a hot rolling mill. All samples were characterised using optical microscopy and image analysis. The volume fraction of retained austenite was ascertained using a heat tinting technique and X-ray diffraction measurements. Room temperature mechanical properties were determined by a tensile test. From this a comprehensive understanding of the structural aspect of the bainite transformation in these types of TRIP steels has been developed. The  results have shown that the final microstructures of thermomechanically processed TRIP steels comprise 50 % of polygonal ferrite, 7 - 12 % of retained austenite, non-carbide bainitic structure and martensite. All steels exhibited a good combination of ultimate tensile strength and total elongation. The microstructure-property examination revealed the relationship between the composition of TRIP steels and their mechanical properties. It has been shown that the addition of Mo to the C-Si-Mn-Nb TRIP steel increases the ultimate tensile strength up to 1020 MPa. The stability of the retained austenite of the Nb-Mo steel was degraded, which led to a decrease in the elongation (24 %). The results have demonstrated that the addition of Al to C-Si-Mn-Nb steel leads to a good combination of strength (∼ 940 MPa) and elongation (∼ 30 %) due to the formation of refined acicular ferrite and granular bainite structure with ∼7 - 8 % of stable retained austenite. Furthermore, it has been found that the addition of Al increases the volume fraction of bainitic ferrite laths. The investigations have shown an interesting result that, in the Nb-Mo-Al steel, Al has a more pronounced effect on the microstructure in comparison with Mo. It has been found that the bainitic structure of the Nb-Mo-Al steel appears to be more granular than in the Nb-Mo steel. Moreover, the volume fraction of the retained austenite increased (12 %) with decreasing bainitic ferrite content. The results have demonstrated that this steel has the best mechanical properties (1100 MPa and 28 % elongation). It has been concluded that the combined effect of Nb, Mo, and Al addition on the dispersion of the bainite, martensite and retained austenite in the ferrite matrix and the morphology of these phases is different than effect of Nb, Mo and Al, separately.

Effect of alloying elements on the microstructure-property relationship in thermomechanically processed C-Mn-Si TRIP steels

The effect of additions of Nb, Al and Mo to Fe-C-Mn-Si TRIP steel on the final microstructure and mechanical properties after simulated thermomechanical processing (TMP) has been studied. The laboratory simulations of discontinuous cooling during TMP were performed using a hot rolling mill. All samples were characterised using optical microscopy and image analysis. The volume fraction of retained austenite was ascertained using a heat tinting technique and X-ray diffraction measurements. Room temperature mechanical properties were determined by a tensile test. From this a comprehensive understanding of the structural aspect of the bainite transformation in these types of TRIP steels has been developed. The results have shown that the final microstructures of thermomechanically processed TRIP steels comprise 50 % of polygonal ferrite, 7 - 12 % of retained austenite, non-carbide bainitic structure and martensite. All steels exhibited a good combination of ultimate tensile strength and total elongation. The microstructure-property examination revealed the relationship between the composition of TRIP steels and their mechanical properties. It has been shown that the addition of Mo to the C-Si-Mn-Nb TRIP steel increases the ultimate tensile strength up to 1020 MPa. The stability of the retained austenite of the Nb-Mo steel was degraded, which led to a decrease in the elongation (24 %). The results have demonstrated that the addition of Al to C-Si-Mn-Nb steel leads to a good combination of strength (∼ 940 MPa) and elongation (∼ 30 %) due to the formation of refined acicular ferrite and granular bainite structure with ∼7 - 8 % of stable retained austenite. Furthermore, it has been found that the addition of Al increases the volume fraction of bainitic ferrite laths. The investigations have shown an interesting result that, in the Nb-Mo-Al steel, Al has a more pronounced effect on the microstructure in comparison with Mo. It has been found that the bainitic structure of the Nb-Mo-Al steel appears to be more granular than in the Nb-Mo steel. Moreover, the volume fraction of the retained austenite increased (12 %) with decreasing bainitic ferrite content. The results have demonstrated that this steel has the best mechanical properties (1100 MPa and 28 % elongation). It has been concluded that the combined effect of Nb, Mo, and Al addition on the dispersion of the bainite, martensite and retained austenite in the ferrite matrix and the morphology of these phases is different than effect of Nb, Mo and Al, separately.

Analytical model of pass-by-pass strain in rod (or bar) rolling and its applications to prediction of austenite grain size

Rod rolling is a process where the deformation state of the workpiece between the work rolls is quite different from the strip rolling process. However, in most microstructure evolution models, the simple area strains (natural logarithm of the area reduction ratio) multiplied by a constant have been used to compute pass-by-pass evolution of austenite grain size (AGS) in rod (or bar) rolling, without any verification. The strains at a given pass play a crucial role in determining the recrystallization behavior (static or dynamic). In this study, an analytical model that calculates the pass-by-pass strain and strain rate in rod rolling has been developed and verified by conducting four-pass (oval–round) bar and plate rolling experiments. Numerical simulations have then been carried out for the four-pass rolling sequence using the area strain model and the new analytical model, focusing on the effect of the method for calculating the strain on the recrystallization behavior and evolution of AGS. The AGS predicted was compared with those obtained from hot torsion tests. It is shown that the analytical model developed in this study is more appropriate in the analysis of bar (or rod) rolling. It was found that the recrystallization behavior and evolution of AGS during this process were influenced significantly by the calculation method for the deformation parameters (strain and strain rate). The pass-by-pass strain obtained from the simple area strain model is inadequate to be used as an input to the equations for recrystallization and AGS evolution under these rolling conditions.

Integrated model for thermo-mechanical controlled process in rod (or bar) rolling

This study presents an integrated model for computing the thermo-mechanical parameters (cross-sectional shape of workpiece, the pass-by-pass strain and strain rate and the temperature variation during rolling and cooling between inter-stands) and metallurgical parameters (recrystallisation behaviour and austenite grain size—AGS), to assess the potential for developing “Thermo-Mechanical Controlled Process” technology in rod (or bar) rolling, which has been a well-known technical terminology in strip (or plate) rolling since 1970s.

The advantage of this model is that metallurgical and mechanical parameters are obtained simultaneously in a short computation time compared with other models. The model has been applied to a rod mill to predict the exit cross-sectional shape, area and AGS per pass by incorporating the equations for AGS evolution being used in strip rolling. At the finishing train of rod mills, the strain rates reach as high as 1000–3000 s−1 and the inter-pass times are around 10–60 ms.

The results show that the proposed model is an efficient tool for evaluating the effects of process-related parameters on product quality and dimensional tolerance of the products in rod (or bar) rolling. The results of the simulation demonstrated that the equation for AGS evolution being used in strip rolling might have limitations when applied directly to rod rolling at a high strain rate.

Local Austenite grain size distribution in hot bar rolling of AISI 4135 Steel

In this paper, the local distribution of austenite grain size (AGS) was experimentally determined by conducting single round-oval and square-diamond pass hot bar rolling experiments of AISI4135 steel. The rolling experiments were carried out using the laboratory mill. The local distribution of AGS was also determined numerically. In order to predict AGS distribution, the AGS evolution model was combined with three dimensional non-isothermal finite element analyses by adopting a modified additivity rule. AGS evolution model was experimentally determined from hot torsion test according to Hodgson's model. The predicted results were in a reasonably good agreement with experimental results.

Eccentricity and hardness control in cold rolling mills with a dynamically constructed neural controller

The main objective of a steel strip rolling process is to produce high quality steel at a desired thickness.  Thickness reduction is the result of the speed difference between the incoming and the outgoing steel strip and the application of the large normal forces via the backup and the work rolls.  Gauge control of a cold rolled steel strip is achieved using the gaugemeter principle that works adequately for the input gauge changes and the strip hardness changes.  However, the compensation of some factors is problematic, for example, eccentricity of the backup rolls.  This cyclic eccentricity effect causes a gauge deviation, but more importantly, a signal is passed to the gap position control so to increase the eccentricity deviation.  Consequently, the required high product tolerances are severely limited by the presence of the roll eccentricity effects.
In this paper a direct model reference adaptive control (MRAC) scheme with dynamically constructed neural controller was used.  The aim here is to find the simplest controller structure capable of achieving an optimal performance.  The stability of the adaptive neural control scheme (i.e. the requirement of persistency of excitation and bounded learning rates) is addressed by using as the inputs to the reference model the plant's state variables.  In such a case, excitation is due to actual plant signals (states) affected by plant disturbances and noise.  In addition, a reference model in the form of a filter with a desired transfer function using Modulus Optimum design was used to ensure variance in the desired dynamic characteristics of the system.  The gradually decreasing learning rate employed by the neural controller in this paper is aimed at eliminating controller instability resulting from over-aggressive control.  The moving target problem (i.e. the difficulty of global neural networks to perfrom several separate computational tasks in closed -loop control) is addressed by the localized architecture of the controller.  The above control scheme and learning algorithm offers a method for automatic discovery of an efficient controller.
The resulting neural controller produces an excellent disturbance rejection in both cases of eccentricity and hardness disturbances, reducing the gauge deviation due to eccentricity disturbance from 33.36% to 4.57% on average, and the gauge deviation due to hardness disturbance from 12.59% to 2.08%.

Misunderstanding John Stuart Mill on science: Paul Feyerabend`s bad influence

Feyerabend, 1975; Feyerabend, 1981 and Feyerabend, 1987 takes J. S. Mill’s On Liberty to support the proliferation of theories in science and to emphasize the fallibility of scientific knowledge. On Liberty, according to Feyerabend, contradicts and overthrows Mill’s major study of science, A System of Logic. Staley (1999) reads the 2 works of Mill as giving complementary accounts of science. The present author rejects these interpretations of Mill, arguing that A System of Logic and On Liberty are mutually compatible since On Liberty concerns the nature of non-scientific knowledge and the methods that Mill believes are appropriate to expanding and assessing that knowledge.

Microstructural evolution of ultrafine grained structure in plain carbon steels through single pass rolling

In the present study, the effect of nominal equivalent strain (between 0 and 1.2), deformation temperature (790– 750°C) and carbon content (0.06 – 0.35%C) was investigated on ferrite grain refinement through dynamic strain induced transformation (DSIT) in plain carbon steels in single pass rolling. The microstructural evolution of the transformation of austenite to ferrite has been evaluated through the thickness of the strip. The results showed a number of important microstructural features as a function of strain, which could be classified into three regions; no DSIT region, DSIT region, and ultrafine ferrite (UFF) grain region. Hence, two critical strains; dynamic strain induced transformation (εC, DSIT) and ultrafine ferrite formation (εC, UFF) were determined. These strains were increased significantly with an increase in carbon content. The critical strain for UFF formation reduced with decrease in deformation temperature. The UFF microstructure consisted of ultrafine, equiaxed ferrite grains (<2 μm) with very fine cementite particles. In the centre of the rolled strip, there was a conventional ferrite– pearlite microstructure, although ferrite grain refinement and the volume fraction of ferrite increased with increase in the nominal equivalent strain.