Zasnova metode za napovedovanje obrabe orodja v odrezovalnih procesih

Žagar, Miha (author), Pušavec, Franci (mentor)

V diplomski nalogi je obravnavan problem napovedovanja obrabe orodij v odrezovalnih procesih, kar je ključno za spremljanje in izboljšanje kakovosti proizvodnje. Predstavljena je metodologija uporabe metode Support Vector Regression (SVR) za natančno napovedovanje obrabe orodij na podlagi večsenzorskih podatkov. Eksperimentalni del naloge vključuje izvedbo testiranj, kjer so bile merjene rezalne in podajalne sile ter obraba orodja. Rezultati so pokazali, da metoda SVR omogoča zanesljive napovedi, kar pripomore k boljšemu razumevanju in spremljanju obrabe orodij med proizvodnim procesom.

• Language: Slovenian
• Keywords: obraba orodja, odrezovalni procesi, strojno učenje, Support Vector Regression, napovedovanje, večsenzorski podatki
• Work type: Final paper
• Typology: 2.11 - Undergraduate Thesis
• Organization: FS - Faculty of Mechanical Engineering
• Year: 2024
• Number of pages: XIII, 57 f.
• PID: 20.500.12556/RUL-160659
• UDC: 621.9:539.375.6:531.78(043.2)
• COBISS.SI-ID: 214969347
• Publication date in RUL: 03.09.2024

ŽAGAR, Miha, 2024, Zasnova metode za napovedovanje obrabe orodja v odrezovalnih procesih [online]. Bachelor’s thesis. [Accessed 26 April 2025]. Retrieved from: https://repozitorij.uni-lj.si/IzpisGradiva.php?lang=eng&id=160659

Secondary language

• Language: English
• Title: Design of a method for predicting tool wear in machining processes

Abstract

This thesis addresses the problem of predicting tool wear in machining processes, which is crucial for monitoring and improving production quality. The methodology of using Support Vector Regression (SVR) to accurately predict tool wear based on multi-sensor data is presented. The experimental part of the thesis includes conducting tests where cutting and feed forces, as well as tool wear, were measured. The results showed that the SVR method provides reliable predictions, contributing to better understanding and monitoring of tool wear during the production process.

• Keywords: toolwear, maschining processes, maschine learning, Support Vector Regression, prediction, multi-sensor data