Data Mining and Statistical Analysis
The group works on the development of data-driven and physically-based modeling strategies and their application to analysis and interpretation of materials data.
At the moment, one of key research topics is the development of physically-based and data-driven models to identify statistically sound correlations between materials chemistry, thermodynamic, microstructure and mechanical data of single crystal Ni- and Co-based super alloys. In comparison to the existing purely phenomenological modeling methods, such hybrid modeling strategies allow to identify the influence of individual physical effects from the considered contribution on selected material properties. Moreover, they are frequently required to accelerate the computer-based design of new materials and alloys, for example, the development of rhenium-free Ni-based alloys with the desirable mechanical properties.
Normalized standardized regression coefficients of alloying elements and their variations in at% for property variation within a full linear regression model.
Due to the high complexity and heterogeneity of materials data, we focus mainly on three key goals:
- the establishment of a well-organised data infrastructure for storage of experimental data and simulation results,
- the development of robust physics-based models of materials properties and their application for thermodynamic calculations and integration into computer-based alloy-by-design framework,
- the development and application of data mining and machine learning methods to identify statistically sound correlations between materials chemistry, microstructure and mechanical data.
Third generation CALPHAD databases from 0K by automated statistical regression analysis.
Moreover, a combination of this modelling technique and the application of statistical design of experiments allow to reduce the number of experimental trials required for the development of new materials and processes.
- Data mining and machine learning
- Automation of data processing
- Physics-based models
- Design of experiments
- Third generation CALPHAD databases: mathematical aspects
Dr. Irina Roslyakova
Department of Scale Bridging Thermodynamic and Kinetic Simulation
Tel: +49 234 32 22605
Fax: +49 234 32 14989
A. Müller, I. Roslyakova, M. Sprenger, P. Git et al. MultOpt++: a fast regression-based model for the development of compositions with high robustness against scatter of element concentrations Modelling and Simulation in Materials Science and Engineering, IOP Publishing, 27, 1-18, (2019)
A. Obaied. Application of machine learning for thermo-physical properties of transition metals Master Thesis, Ruhr-Universität Bochum (2018)
J. Yuxun, S. Zomorodpoosh, I. Roslyakova, L. Zhang. Thermodynamic re-assessment of binary Cr-Nb system down to 0 K CALPHAD, Elsevier Ltd., 62, 109-118, (2018)
M. Almodallaleh. Anwendungen von automatisierter Datenverarbeitung auf Doppelscherversuche von Nickelbasissuperlegierung Master Thesis, Ruhr-University Bochum (2018)
I. Roslyakova. Modeling thermodynamical properties by segmented non-linear regression PhD Thesis, ICAMS, Ruhr-Universität Bochum (2017)