Laser welding of new austenitic cryogenic corrosion-resistant steels alloyed with nitrogen

Abstract

The article proposes a method for improving the thermal performance of walking beam reheating furnaces using a statistical mathematical model of metal heating. The main part of the study refers to the creation of a metal heating model based on a comprehensive analysis of 15 industrial experiments carried out on the Russian continuous furnaces. The first stage of statistical model creation consists of factors’ selection, construction of regression model, correlation analysis and assessment of the variables’ significance, adjustment of factors and obtaining regression equations. As a result, a quasi-dynamic statistical mathematical model of a five-zone walking beam reheating furnace has been created and adapted, that actually makes it possible to trace the heating process in dynamics. The adaptation of the statistical model and error calculation has been carried out using the results of industrial experiments on the investigated furnaces of mills 2000 and 5000. The article contains graphs comparing real temperatures and temperatures calculated on the basis of mathematical and statistical models for one of the experiments. The second stage of quasidynamic statistical model creation has shown that temperature of the combustion products of neighboring zones is a significant factor, that enhances calculation accuracy by 1-3%. The main conclusions are formulated based on the results of the research done. For the first time in metallurgical practice, a quasi-dynamic statistical model has been developed and adapted which describes the process of metal heating in a walking beam reheating furnaces. Since the regression function is defined, interpreted and justified, and the assessment of the accuracy of the regression analysis meets the requirements, it can be assumed that the model and predicted values have sufficient reliability.