Flank wear modeling in high speed hard end milling using integrated approach of Monte Carlo simulation method and Taguchi design
In high speed cutting of hard materials, the wear rate will be very difficult to predict due to the fast and sever changing in the cutting zone. Therefore, using the traditional methods in predicting the output responses will be not the correct options. One of the effective alternatives is by us...
Saved in:
Main Authors: | , |
---|---|
Format: | Conference or Workshop Item |
Language: | English English English |
Published: |
Institute of Electrical and Electronics Engineers Inc.
2019
|
Subjects: | |
Online Access: | http://irep.iium.edu.my/79666/20/79666_Flank%20Wear%20Modeling%20in%20High%20Speed_complete_new.pdf http://irep.iium.edu.my/79666/8/79666_Flank%20Wear%20Modeling%20in%20High%20Speed%20Hard%20End%20Milling_scopus.pdf http://irep.iium.edu.my/79666/9/79666_Flank%20Wear%20Modeling%20in%20High%20Speed%20Hard%20End%20Milling_wos.pdf http://irep.iium.edu.my/79666/ https://icecta.aurak.ac.ae/wp-content/uploads/2019/11/Icecta-Program-2019-3.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | In high speed cutting of hard materials, the wear
rate will be very difficult to predict due to the fast and sever
changing in the cutting zone. Therefore, using the traditional
methods in predicting the output responses will be not the
correct options. One of the effective alternatives is by using
artificial intelligent approach. The current work presents the
simulation of flank wear rate in high-speed hard end milling of
AISI H13 hardened steel using an integrated approach of using:
Monte Carlo (MC) simulation method based on Taguchi design.
An experimental investigation was carried out using coated
carbide tools to run a set of experiments using Taguchi design
(L9) with three input factors at three levels in the following
design boundary: cutting speeds (352-452 m/min), feed rate
(0.01-0.05 m/rev), and depth of cut of (0.2-0.5) mm. Each
experiment was repeated twice using three inserts. The results
was used to create 1000 run simulated from 135 experimental
reading. A new model was developed using JMP software. The
results were analyzed statistically and indicate that even with
the complexity of the process, the neural network technique was
found to be adequate in predicting and simulating the flank
wear length. |
---|