PCA based hybrid Taguchi philosophy for optimization of multiple responses in EDM

The present study is aimed at a multi-response optimization problem by applying Principal Component Analysis (PCA) combined with Taguchi method. The investigation has been carried out through a case study in Electric Discharge Machining (EDM) of D2 steel by using copper, brass and Direct Metal Laser...

Full description

Saved in:
Bibliographic Details
Published in:Sadhana (Bangalore) 2019, Vol.44 (1), p.1-9, Article 2
Main Authors: Mohanty, S D, Mahapatra, S S, Mohanty, R C
Format: Article
Language:English
Subjects:
EDM electrodes
Electric discharge machining
Engineering
Laser sintering
Material removal rate (machining)
Mathematical analysis
Multiple objective analysis
Optimization
Performance indices
Principal components analysis
Process parameters
Surface roughness
Taguchi methods
Tool wear
Variance analysis
Wear rate
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:The present study is aimed at a multi-response optimization problem by applying Principal Component Analysis (PCA) combined with Taguchi method. The investigation has been carried out through a case study in Electric Discharge Machining (EDM) of D2 steel by using copper, brass and Direct Metal Laser Sintered (DMLS) electrode produced by direct metal laser sintering using Directmetal20. The research work has been carried out to evaluate the best parametric combination which could fulfill multiple responses like lower Tool Wear Rate (TWR), higher Material Removal Rate (MRR) and lower Surface Roughness (R a ). Unlike the use of single-objective optimization problem in traditional Taguchi method, a hybrid Taguchi method has been developed in combination with PCA to solve multi-objective problem. Taguchi method assumes that the quality characteristics should be uncorrelated or independent which is not always fulfilled in actual condition. PCA is applied to remove response correlation and to calculate independent (uncorrelated) quality indices known as principal components. These principal components combined with weighted principal component analysis (WPCA) are used to calculate overall quality index denoted as Multi-response Performance Index (MPI). This investigation combines WPCA and Taguchi method for forecasting optimal setting. The predicted result by this method was validated through confirmatory test proving the efficacy of the process. Out of five input process parameters considered, tool electrode has been found to be the most significant factors through the Analysis of Variance (ANOVA).
ISSN:0256-2499
0973-7677
DOI:10.1007/s12046-018-0982-z