EUROPEAN CONGRESS AND EXHIBITION ON ADVANCED MATERIALS AND PROCESSES - EUROMAT 2021

During the last decades, the design of thermal treatments has become an issue of importance due to the higher mechanical requirements imposed in the market, which implies microstructural optimizations. Although the most precise way of designing a heat treatment for a specific steel involves the experimental estimation of the steel critical temperatures, it is undeniable that this estimation is very time and resource-consuming, which, eventually, means expensive. In this context, the development of models for the prediction of such critical temperatures can save money. The authors of this work have reviewed a wide range of models found in the literature and compared their results with a large experimental database. The comparison has highlighted the necessity of accurate models which include the effect of a wider amount of chemical elements while keeping a low error. In this work, such a need has been solved by taking advantage of the most recent progress in machine learning algorithms, developing models that present better results than the ones in the literature at the same time that they include a wider range of elements.