Title: Artificial neural network modelling and analysis of carbon nanopowder mixed micro wire electro discharge machining of gold coated doped silicon

Authors: Sams Jarin; Tanveer Saleh

Addresses: Department of Mechatronics Engineering, International Islamic University Malaysia, 50728 Kuala Lumpur, Malaysia ' Department of Mechatronics Engineering, International Islamic University Malaysia, 50728 Kuala Lumpur, Malaysia

Abstract: This research aims to machine gold coated doped Silicon (Si) wafer using micro wire electro discharge machining or uWEDM process in the presence of carbon nanopowder mixed dielectric oil. Effects of uWEDM parameters namely voltage, capacitance, powder concentration and coating thickness on average surface roughness (ASR), material removal rate (MRR) and spark gap (SG) have been analysed. All the three output parameters, i.e. material removal rate (MRR), spark gap (SG) and average surface roughness (ASR) were found to follow the parabolic trend with the variation of the nanopowder concentration. SG and ASR both were observed to be increased with the coating thickness. However, MRR was decreased with the same. An Artificial neural network based technique was used to model various responses. The overall model prediction was found to be in good agreement (average error less than 10%) with the experimental results for the corresponding input process parameters.

Keywords: uWEDM; nanopowder; silicon; artificial neural network; ANN; gold; doped.

DOI: 10.1504/IJMATEI.2019.103614

International Journal of Materials Engineering Innovation, 2019 Vol.10 No.4, pp.346 - 363

Received: 05 Feb 2019
Accepted: 05 Jun 2019
Published online: 13 Nov 2019 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article