Desarrollo de un metodo numerico y validacion experimental para el calculo del campo de temperaturas en la soldadura de un acero inoxidable y su aplicacion en edificacion.

摘要

Use of stainless steels for building has increased in a considerable way from 1990 and the applications of this material have been seen diversified in relevant construction projects. Many of these uses involve structures, facades and components of facades, components of different systems, urban elements, sculptures, etc. Arc welding process is, generally speaking, the most used joint technique for stainless steels, even for thin sheets applications. The temperature-time cycles during the welding operation have a decisive influence on the mechanical properties of the welded joint as a consequence of the thermal action on the material.This doctoral thesis develops a finite difference method to evaluate the temperature field in the heat affected zone in butt welding of AISI 304 stainless steel sheet by GTAW process. A computer program has been developed and it has been implemented by Visual Basic for Applications (VBA) in MS-Excel spreadsheet. The modelled experiments with the numerical application foresee the thermal behaviour of arc welding processes. One of the most relevant contributions in the simulation field is the possibility of knowing the effect of every variable involved in the process as well as establishing the weight of them. This leads to a better knowledge of the phenomenon, which is used to improve and to control the different welding parameters. That is, if a variable effect is known and it is suitably treated, the process output variables may be controlled.It has been developed an experimental methodology to validate the mathematical model by the measurement of the temperature field in an area, close to the weld bead. The methodology is applied to a stainless steel sheet with a thickness lower than 3 mm, although may be used for other steels and welding processes as MIG/MAG and SMAW. The data which has been obtained with these tests have been used to validate the results that have been calculated by the finite difference numerical method. The mathematical model adjustment has been carried out taking into account the experimental results. The differences found between the experimental and theoretical curves are due to the convection and radiation heat losses, which have not been considered in the simulation model. With this simple model, the designer will be able to calculate the thermal cycles that take place in the process as well as to predict the temperature field in the proximity of the weld bead.As an application, it has been employed the numerical simulation model to calculate the t8/5 parameter values and this one has been optimized by a particular consideration of the response surface methodology.