Thesis supervisor: Jurij Sidor
co-supervisor: Péter Kovács
Location of studies (in Hungarian): Eötvös Loránd University, Faculty of Informatics Abbreviation of location of studies: ELTE
Description of the research topic:
The processing of materials involves a complex thermomechanical chain. Each element of processing is strongly correlated to the internal structure of materials and defines the final properties and performance. The traditional computational algorithms have evolved to the level, that they enable to capture the evolution of both microstructure and related properties, however, in many instances these numerical approaches involve a number of model parameters, which need to be defined for each material type and each processing route separately. Another challenge is the amount of data that can be obtained during measurements, which is usually not enough for the direct application of pure data-driven artificial intelligence algorithms. Obtaining a generic knowledge about the development of structural elements on various scales such as microscopic and nanoscale requires involvement of knowledge-augmented deep-learning algorithms that combine the advantages of computational techniques derived from mathematical models and data-based machine learning.
In view of this, the major goal of the PhD work is to explore the limits of existing and development of new algorithms enabling finding the correlation between the processing parameters and the evolution of characteristic microstructure and substructure during processing. To this end, it is necessary to develop knowledge-augmented neural networks in which domain knowledge can be incorporated into the network architecture, the loss function, and the feature space of the data. This way, not only accurate, but also consistent estimates can be given in accordance with physical laws. The machine learning algorithms will be trained by employing experimental data obtained after deformation and heat treatment processes. The results gained from virtual processing will be used for the modification of already existing technologies as well as development of new ones.
Further requirements: MSc diploma in one of the following fields: informatics, materials engineering, civil engineering, physics, mechanical engineering or related areas.
Number of students who can be accepted: 1
Deadline for application: 2024-05-31
2024. IV. 17. ODT ülés Az ODT következő ülésére 2024. június 14-én, pénteken 10.00 órakor kerül sor a Semmelweis Egyetem Szenátusi termében (Bp. Üllői út 26. I. emelet).
Login
