The development of rolling mathematical models began more than 80 years ago, with the conclusion of the empirical Ekelund's formula for the calculation of the rolling load. Von Karman and Orowan's studies, developed in the following decades, provided the theoretical and mathematical basis for this molding process. On the other hand, they equally revealed the huge associated mathematical and physical complexity, which could only be solved with great difficulty in view of the rare calculation resources available at the time. The next step was the development of simplified models to make the calculations easier, overcome the lack of metallurgic data and allow their routine application in the industry. Such simplifications, however, led, many times, to great deviations between the results calculated and the results observed, compromising the models reliability.
This way, their mandatory application requires correction through empirical learning schemes. The digital revolution which took place in the last decades, associated to a systematic research work aiming at collecting the process' basic metallurgic data and at the instrumentation progress, is increasingly contributing to improve the precision models.
The larger computer's capacity available at low costs has turned the use of more complex and accurate models, as well as the ones based on the finite elements method, into routine. Today it is also possible to minimize the inconveniences arising from occasional lack of data or process inconsistency through artificial intelligence techniques, such as neural networks and fuzzy logic. On the other hand, such solutions also created new problems.
The transfer of more complex and accurate calculation methods for the industrial environment is still a problem, once the process speed available is still not enough for its execution within suitable time. Besides, the efficient application of such new methods requires highly trained personnel, with a wide experience in their use. Experience shows that the best way to respond to this challenge lies in an effective association between the industry and the academy which allows the conclusions of their competencies in a harmonious and productive way. |
