Multi-Objective Optimization in Dry Turning of Hardened 20MnCr5 Using Random Forest Model

Dry machining has become a viable alternative to traditional machining processes, given the various drawbacks associated with the use of cutting fluids. In this work, 20MnCr5 has been selected as workpiece material to perform a dry turning operation after it had been case-hardened up to 51 HRc. It i...

Full description

Saved in:
Bibliographic Details
Published in:Journal of physics. Conference series 2024-10, Vol.2837 (1), p.12073
Main Authors: Venkata Ramana, M., Prasad, M.V.R.D., Yugeshwar, C.
Format: Article
Language:English
Subjects:
Accuracy
ANOVA
Case hardening
Cutting fluids
Cutting parameters
Cutting speed
Cutting wear
Dry machining
Dry turning
Grey relational analysis
Machine learning
Material removal rate (machining)
Multi objective optimization
Multiple objective analysis
Optimization
Random Forest model
Surface roughness
Taguchi technique
Tool wear
Turning (machining)
Wear rate
Workpieces
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Dry machining has become a viable alternative to traditional machining processes, given the various drawbacks associated with the use of cutting fluids. In this work, 20MnCr5 has been selected as workpiece material to perform a dry turning operation after it had been case-hardened up to 51 HRc. It is crucial to find the behaviour of the hardened material under dry machining as the usage industries are going toward dry machining. For the experimentation PVD coated TH1000 insert grade is utilized. Depth of cut (DoC), cutting speed (V c ) and feed (f) were the input parameters. Surface roughness (R a ), tool wear (T w ), material removal rate (MRR) and power consumption (P c ) were the performance characteristics examined after the machining process. Multi-objective optimization was performed using grey relational analysis. In multi-objective optimization, the optimal condition was achieved at 120 m/min cutting speed, 0.1 mm depth of cut and 0.1 mm/rev feed. All the machine learning models developed had an accuracy of 90%. The predicted values from machine learning models were utilised to determine the optimal condition in multi-objective optimization. The results obtained for multi-objective optimization from the random forest model were also 0.1mm/rev feed, 120m/min cutting speed and 0.1 mm depth of cut. Feed was the factor that had most influenced by 14.78% on multi-objective optimization. It was observed that the models developed had high accuracy of 86.7% and there was significant harmony between actual and predicted values.
ISSN:1742-6588
1742-6596
DOI:10.1088/1742-6596/2837/1/012073