A machine learning-aided multisensor system for thermomechanical measurements and in-depth analysis of the surface layer in aircraft alloys

Polish Metrology II

Leader: Rzeszow University of Technology

Partner: Lublin University of Technology

Project title: A machine learning-aided multisensor system for thermomechanical measurements and in-depth analysis of the surface layer in aircraft alloys

Agreement number: PM-II/SP/0040/2024/02

Project implementation period: 01.03.2024 - 28.02.2026

Principal Investigator (Rzeszow University of Technology): dr hab. inż. Witold Habrat, prof. PRz

Principal Co-investigator  (Lublin University of Technology): dr inż. Magdalena Zawada - Michałowska

Project value: 968 000,00 PLN

Funds granted for Lublin University of Technology: 484 000,00 PLN

Funds granted for Rzeszow University of Technology: 484 000,00 PLN

Abstract: The aim of the project implemented in the discipline of mechanical engineering is to develop measurement procedures for an integrated system of multisensory measurements of the impact of thermo-mechanical interactions in the cutting processes of light alloys used in aerospace technology on the condition of the technological surface layer. In the case of analyzing the impact of the measured input quantities (and their interactions) on the indicators the effectiveness of loss shaping, machine learning algorithms will be used to identify the relationships between factors.

A standard approach to analyzing the cutting process is to determine the influence of cutting parameters and other technological variables (cutting tool geometry, cutting conditions, workpiece material, cutting edge material, etc.) on machinability indices. In the case of aircraft parts, especially critical elements responsible for flight safety, in addition to meeting the design requirements in terms of dimensional and shape accuracy and the quality of the machined surface, it is also required to determine the impact of the cutting process on unmachined zones. Such a comprehensive view of research requires a systematic approach to measurement along with the development of procedures that facilitate the comparison of results. Taking into account the need to establish a conventional measure in the field of material testing (e.g. changes in the morphology of the microstructure of the surface layer), identifying the magnitude of plastic and thermal interactions, it is planned to use machine learning algorithms, which will allow the development of a tool for comparing the condition of the surface layer and blade wear.