MODELLING THE CRYSTALLIZER’S MOLD STATE DEPENDING ON ITS WEAR

Abstract

This article is designed to determine the life of technical objects, in particular molds. Specifying lifetime of technical objects are followed by an explanation of standards and in particular standard EN 61649:2008 Weibull analysis. Further detail is given technical characteristics of the molds. Within the two models have been developed mold life - the first based on analytical expressions of the degradation mechanism transferred to the shape of the marginal probability of failure with the application of Weibull distribution, the second model using artificial neural networks based on the frequency dependence of individual brands of cast steel and mold profile of the average value measured at each system levels MKL 100/420. Reliability and maintenance control systems gives us an analysis of the maintenance actions, failure analysis, costs analysis spent to made actions and spare parts. It gives the answers to questions, which would be without a complex look over the data derived hardly – if the given crystallizer’s desks can be used in operation or if is necessary to make its renovation. On history basis of previously actions or another historical data the maintenance control system gives sufficient information for correct decision make. Thanks to recently made maintenance data collection the base for maintenance planning is improving. Introduction of reliability control system and maintenance further help to increase availability and serviceability of devices from the reason of lowering unplanned time delays, breakdowns and increase device’s service life.