Vliv tepelných procesů na kvalitu plynule litých bram a návrh predikce vad

Abstract

The dissertation thesis summarizes the basic analytical and empirical findings from analysis of dependences of the influence of thermal processes continuous casting machine for steel (hereinafter CCM) on quality of slabs, and thus also the influence of thermal processes on the quality of final products rolled from slabs, mainly steel sheets. The course of thermal processes at continuous steel casting has a principal influence on the quality of slabs and thus also on the entire economics of production in the steel works, as well as in the whole metallurgical plant. The assessment of the quality of steel slabs on CCM is an inseparable part of the information system of the metallurgical plant. This assessment woks on the basis of collection and storage of the data that are necessary for an efficient assessment of dependencies between the measured and qualitative quantities. The data acquired by measurement at sequence casting, by chemical analysis, as well as data entered by the steel shop operators are stored and archived. In this manner a time line of huge amount of data is continuously being generated. The thesis describes the proposal of the concept of efficient storage of these data and of their use both for operational control of production and for statistic evaluation of long-dated data and quality optimisation. Such approach requires creation of super-structural software for standard information system of the steel shop. The original software system LITIOS, which was developed and implemented on CCM in the steelmaking company EVRAZ Vitkovice Steel, a.s., can serve as an example of such software. This proprietary software gives to technologists on the slab CCM a possibility not only to analyse the history of casting, but also to perform long-term monitoring and optimisation of production. Practical part of the thesis presents analysis of mathematical or statistical methods for the assessment of slab casting process, which enable also prediction of slab quality, as well as quality of the final rolled products. This analysis is substantiated by statistical data analysis of concrete data. The author has used his experience gained at processing of extensive data files from continuous casting machines at metallurgical plants Třinecké Železárny, a.s., US Steel Košice and particularly at EVRAZ Vítkovice Steel, a.s. Experience and knowledge gained by technologies during many years at casting of various grades of steel together with statistic analyses served as a basis for designing and implementation of the original system for prediction of slab quality.