Mamdani Fuzzy Inference System Modeling to Predict Surface Roughness in Laser Machining

Sivarao, Subramonian (2009) Mamdani Fuzzy Inference System Modeling to Predict Surface Roughness in Laser Machining. International Journal of Intelligent Information Technology Application, 2 (1). pp. 12-18. ISSN 1999-2459

[img] PDF
Restricted to Registered users only

Download (1MB) | Request a copy


The world of manufacturing has shifted its level to the era of space age machining. The purpose of this investigation is to develop Fuzzy based Graphical User Interface (GUI) for modeling of laser machining conditions. The developed fuzzy based GUI is expected to overcome the major problems faced by most of the manufacturing industries nowadays with the increased number controllable parameters and the lack of expertise to operate the machine. Investigations were begun by screening for the significant parameters before the design for GUI is made. Then, the GUI for Fuzzy based modeling has been developed using GUIDE Toolbox and Fuzzy Toolbox. The fuzzy variables were also analyzed before finalizing the significant of its variables. The GUI developed has been programmed to interact with fuzzy variables in order to model the laser processing cut quality of two different thicknesses, 2.5 and 5 mm. The models were then compared for their statistical validation by Root Mean Square Error (RMSE) values. Few models with best and optimized variables were taken as prediction models, where their respective outputs were analyzed and compared based on percentage error for 128 data sets to validate the models. The best developed model was then recommended to the pressure vessel manufacturing industry to further reduce the production cost and enhance it end product quality.

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 15:20
Last Modified: 28 May 2015 04:01
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item