Wire-EDM process parameters and optimization

Umesh Kumar Vates

ISBN 10: 6202305789 / ISBN 13: 9786202305785

Published by Scholars' Press, 2018

About this Item

nach der Bestellung gedruckt Neuware - Printed after ordering - The present work is devoted to the development of ANN models for the analysis of the WEDM process applied to an actual industrial surface quality of components of mould and dies. Four layered structure (including two hidden layers, input and output layer) with back propagation for ANN model is selected and two fold cross-over data approach has been used to develop two different sets of models (S1 & S2) for each experimental set of material rods. Extensive computational results (predicted output values) have been obtained by using training, validation and testing programmed in 'MATLAB R2010a'. Prediction error statistics and correlations coefficient testing carried out between experimental and predicted outputs Ra as well as MRR to evaluate the best modeling result. The best among the proposed ANN models has been used to optimize the process parameter (Ra) with MRR by the effort of best input parametric combination.

Seller Inventory # 9786202305785

Bibliographic Details

• Title: Wire-EDM process parameters and optimization
• Publisher: Scholars' Press
• Publication Date: 2018
• Language: English
• ISBN 10: 6202305789
• ISBN 13: 9786202305785
• Binding: Taschenbuch
• Condition: Neu
• Dimensions: 220W x 150H x 18D millimeters 220 millimeters width by 150 millimeters height by 18 millimeters depth
• Weight: 453 grams

About this title

Synopsis

The present work is devoted to the development of ANN models for the analysis of the WEDM process applied to an actual industrial surface quality of components of mould and dies. Four layered structure (including two hidden layers, input and output layer) with back propagation for ANN model is selected and two fold cross-over data approach has been used to develop two different sets of models (S1 & S2) for each experimental set of material rods. Extensive computational results (predicted output values) have been obtained by using training, validation and testing programmed in “MATLAB R2010a”. Prediction error statistics and correlations coefficient testing carried out between experimental and predicted outputs Ra as well as MRR to evaluate the best modeling result. The best among the proposed ANN models has been used to optimize the process parameter (Ra) with MRR by the effort of best input parametric combination.

About the Author

Dr. Umesh Kumar VatesPhD, IIT Dhanbad, India.