Analysis of cutting forces in helical ball end milling process using machine learning

In this paper, variation of cutting forces exerted on tool with respect to cutting parameters for ball end milling process is analysed using deep neural network. A neural network was fabricated to predict cutting forces in all three orthogonal directions for a given set of cutting parameters. Analys...

Full description

Saved in:
Bibliographic Details
Published in:Materials today : proceedings 2021, Vol.46, p.9275-9280
Main Authors: Balasubramanian, Ananth Narayana, Yadav, Naman, Tiwari, Asim
Format: Article
Language:English
Subjects:
Ball end mill
Machine learning
Milling
Neural network
Shell scripting
Titanium alloy
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, variation of cutting forces exerted on tool with respect to cutting parameters for ball end milling process is analysed using deep neural network. A neural network was fabricated to predict cutting forces in all three orthogonal directions for a given set of cutting parameters. Analysis was done by varying axial depth of cut and feed-rate while keeping all other parameters constant, because the variation of cutting forces was most significant for these two parameters. For generating data CutPro simulation software was used along with shell scripts to automatically vary the parameters and simulate. Finally, after proper tuning of hyperparameters of the neural network, maximum percent deviation of predicted values over test dataset was brought down to less than 1 percent.
ISSN:2214-7853
2214-7853
DOI:10.1016/j.matpr.2020.02.098