Thesis supervisor: Mátyás Andó
co-supervisor: Béla János Szekeres
Location of studies (in Hungarian): Szombathely, Budapest Abbreviation of location of studies: ELTE
Description of the research topic:
CNC machines still make up the majority of production machines. Today the unattendence (light-off) machining become more and more important, but the basis of the cutting technologies is unchanged. Tool breakage and wear are natural phenomenon under machining, therefore automated tool monitoring become key question in modern production cells. Artificial intelligence gives us new possibility to monitoring the tool condition based on camera images. This solution increase the production reliability and decrease the machining time compare with the traditional toll setting methods. Moreover it can use to inspect the workpiece which helps to identify the scrap on time, and avoid the unnecessary machining.
Objectives: The aim of the research is to develop innovative deep learning algorithms and models capable of identifying the type of tool, its size, as well as tool breakage and edge fracture. The completed image also creates the opportunity to check the suitability of the raw-material or even to monitor the effects of intermediate steps in the workflow during tool changes. The model must be applicable for use with new machines or in cases of different camera angles.
Current devices: CNC milling machine (with integrated camera), CNC lathe (with integrated camera), different tools and toolholders
Applied technologies: Python, PyTorch/TensorFlow
Required language skills: angol Further requirements: The following skills are needed:
• Data collection from different sources
• Image processing
• Development of artificial neural networks
• Technology based testing and validation
• Creating real time control system with integrated CNN
Number of students who can be accepted: 2
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
