No abstract

No abstract

Numerical models of solidification and cooling of continuously cast billets or blooms are used both in research and in operational conditions to predict solid shell thickness, metallurgical length, solidification rate etc. The numerical model must be verified according to real values of quantities. Although several different quantities can be used to verify the model, most often the models are verified by comparing the calculated and measured surface temperatures of the strand in the secondary and tertiary cooling zones.The casting process is influenced by a number of known and hidden parameters, often time-varying, which are reflected in the measured surface temperatures, but which cannot be incorporated into the model due to a lack of information to define the exact boundary conditions. For the purposes of model verification, it is therefore necessary to revise the measured data. It is not enough to use only mathematical methods to process data without knowledge of the casting process, because uncertainties and temperature fluctuations have different and often difficult to detect causes. The article deals with sources of temperature uncertainties and fluctuations and methods of extraction of relevant values from measured signals.

No abstract

When describing the behaviour and modelling of real systems, which are characterized by considerable complexity, great difficulty, and often the impossibility of their formal mathematical description, and whose operational monitoring and measurement are difficult, conventional analytical–statistical models run into the limits of their use. The application of these models leads to necessary simplifications, which cause insufficient adequacy of the resulting mathematical description. In such cases, it is appropriate for modelling to use the methods brought by a new scientific discipline—artificial intelligence. Artificial intelligence provides very promising tools for describing and controlling complex systems. The method of neural networks was chosen for the analysis of the lifetime of the teeming ladle. Artificial neural networks are mathematical models that approximate non-linear functions of an arbitrary waveform. The advantage of neural networks is their ability to generalize the dependencies between individual quantities by learning the presented patterns. This property of a neural network is referred to as generalization. Their use is suitable for processing complex problems where the dependencies between individual quantities are not exactly known.