Modelling of CO2 Laser Materials Processing by Networked Neuro-Dimension Fuzzy Intelligent System

Sivarao, Subramonian and S., Thiru and Jusoff, Kamaruzzaman and Azizah, Shaaban and Mariana, Yusoff and Jano, Zanariah and Yuhazri, Yaakob and HASOLOAN , HAERY IAN PIETER and Abu Bakar, Mohd Hadzley and Raja, Izamshah and Hussein, Nur Izan Syahriah and Mohd Amran, Md Ali and Taufik, , and Wahyono Sapto, Widodo and Tan, CheeFai and Sivakumar, Dhar Malingam (2013) Modelling of CO2 Laser Materials Processing by Networked Neuro-Dimension Fuzzy Intelligent System. Australian Journal of Basic and Applied Sciences, 7 (3). pp. 35-45. ISSN 1991-8178

[img] PDF
Restricted to Registered users only

Download (369kB) | Request a copy


The usage of non-contact machine tools in metal cutting industries is increasing exponentially in recent times due to the advent of super hard work materials. The fiercely competitive manufacturing industries find the traditional trial-and-error method, not only challenging and time consuming but also economically nonviable in acquiring superior quality of end products produced from such materials. This paper presents a more accurate prediction method for surface roughness values by using Networked Fuzzy Intelligent System (N-FIS) in combination with the commercial software tools, Matlab and Neuro-Dimension Network. Experiments were conducted on manganese molybdenum pressure vessel plates to validate the robustness of the developed model in forecasting the surface roughness values. The predicted results were found to be reasonably accurate with an error percentage well below 10% for almost all runs at the prediction stage. On the other hand, the experimental validation results for selected critical runs were found to be extremely promising with results of an accuracy of more than 90%. It was concluded that the setting of proper network algorithm and rules, can actually help the industry to perform laser machining better by saving huge waste materials by employing the approach presented in this paper as compared to the traditional trial-and-error method.

Item Type: Article
Subjects: T Technology > TJ Mechanical engineering and machinery
Divisions: Faculty of Manufacturing Engineering > Department of Manufacturing Process
Depositing User: Assoc. Pror. Ir. Dr. Sivarao Subramonian
Date Deposited: 13 Aug 2013 16:44
Last Modified: 28 May 2015 04:02
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item