Electric arc furnaces are intended for the recycling of steel scrap. In the recycling process, the final melt temperature is one of the most important quantities. Due to the nature of the melting process, continuous measurement of the melt temperature is impossible and is performed only before the melt is tapped, in order to check whether the melt temperature is within the prescribed interval. Disposable measuring probes are used to perform the measurements. During the temperature measurement, the furnace must be switched off, which contributes to a longer recycling time, unnecessary energy losses and consequently lower efficiency.
The thesis presents the development of a melt temperature model in an electric arc furnace using a fuzzy approach. The model includes all influencing factors measured on an electric arc furnace. The proposed melt temperature model is intended for implementation on an electric arc furnace, where it will operate in parallel with the recycling process and offer support to electric arc furnace operators on steel temperature. This will reduce the required number of melt temperature measurements, which shortens the recycling time of steel and consequently increases plant productivity. The fuzzy temperature model has proven to be reliable and suitable for use on an electric arc furnace provided that the first temperature measurement is correct and that the steel scrap is completely melted by the first melt temperature measurement.
The fuzzy temperature model is not tied to the specific properties of the electric arc furnace, but only to the measurements performed on the furnace. This allows it to be quickly and easily transferred to other furnace designs. The approaches to fuzzy modelling indicated in the thesis are not limited to estimating melt temperature, as it can be easily extended to other areas.
|
