(6) Metallurgical Integration into Level 2 Model - Online
For some less computing-intensive calculations, the metallurgical and temperature requirements could be integrated into the Level 2 model logics.
The design of the particular rolling passes to produce an ordered set of plates would be achieved by such a metallurgical level 2. Level 2 models use a heuristic or trial-and-error approach to searching for a set of rolling passes including roll gap, quench, speed and roll shape settings. In this search, a metallurgical level 2 with a complete set of models would calculate the probable strength, toughness, flatness and dimensions for a pass design. The calculations would project the grain size, shape, phases present and the strain conditions for each rolling pass. The final projection would be compared to the target properties for the ordered set of plates. Then the gap, quench, speed and roll shape settings would be modified to improve the projections, until a final set of recommendations were complete. In today?s computing level, such an approach is possible but should be carefully tested.
The data recorded at the mill would be stored in the database and analyzed to refine the constants in the models. A systematic and statistically valid approach should be used to refine the constants. The performance of the model in predicting forces, torques, temperature, yield strength, ultimate strength and dimensional uniformity must be reported, and a process engineers should evaluate the reports. The use of the metallurgical models in the design of the rolling passes would provide the maximum toughness with a more accurate achievement of the final target properties without unnecessary loss of steel or processing time.
(7) Property Variations
In an investigation of microstructure and steel properties in various regions of the plate, clear and persistent variations occurred along the length, width and thickness of the steel. This would be expected from the observed variations in processing and the known metallurgical reactions that occurred from the processing. The variations help to explain the difficulty in reaching the target mechanical properties for an order. The microstructure through the thickness of the steel showed that the grains at the center were larger and less flat than the grains on the surface. This would be expected from the metallurgical slip lines of flow during plastic deformation. Depending on the roll diameter and the draft during rolling, the resulting grain nucleation and growth are predictable. Even the variation in the retained strain at the center after the delays between passes could be predicted. None of these processes were included in typical Level 2 models, although these metallurgical processes have a large influence on the mechanical properties of the steel, particularly the impact properties critical in X grades.
There were also observed differences in the microstructure and the temperatures of the steel across the width. Near the edges of the steel, though the temperatures is commonly about 100˚C colder during rolling and quenching., the Level 2 models do not track this difference or the effects of the temperature on the flow stress and the percentage of plastic deformation near the edge. Most models do track the variations in the forces across the work rolls, but do not model the metallurgical results of these differences. Down the length, even short plates had changes in temperatures, changes in flow stress, and subsequent changes in microstructure along the length, particularly for X grades. As mentioned earlier, when rolling steel from coil furnaces, the temperature differences can usually be over 100˚C, and thus the metallurgical effects are significant. Most Level 2 models track the temperatures at the ends of the steel but do not taken into account the pattern of variations from end to end, that affects grain size, phases present and retained strain in the steel. These details are critical to predicting the properties of the steel and to maximizing the utilization of the rolling mill.
A high-quality Level 2 should cover all the above-mentioned issues, for all kinds of purposes including improving rolled steel properties.
(8) Roughing and Finishing Passes Specific
Due to different constrains in roughing and finishing passes, different approaches should apply. Roughing pass is mainly under torque limitation. Decision should be made on the temperature region (above or below the recrystallization temperature) and/or the thickness range (e.g. roughly one half of the slab thickness) in which high reductions should apply. Finishing passes should mainly focus on balancing shape and property. Early finishing passes are under force limitation, while in the last pass or the last two passes, the force and torque limits are not limiting factors any more for the draft scheduling. In addition, high draft and speed help increase temperature that is often a concern particularly in the last one or two passes.
Metal Pass has recently developed several dozens constrains and approaches for the Level 2 model to create high-quality draft schedule aimed at improving rolled product properties. Those can be specified based on each individual plant practice and specific product requirement.
