Genetic algorithm-based drilling burr minimization using adaptive neuro-fuzzy inference system and support vector regression. (April 2020)

• Record Type: Journal Article
• Authors: Mondal, Nripen; Mandal, Madhab Chandra; Dey, Bishal; Das, Santanu
• Is Part Of: Proceedings of the Institution of Mechanical Engineers. Volume 234:Number 5(2020)
• Page Start: 956
• Page End: 968
• Publication Date: 2020-04
• Subjects: Burr height -- burr thickness -- optimization -- adaptive neuro-fuzzy inference system -- support vector regression -- genetic algorithm; Mechanical engineering -- Periodicals; Engineering -- Management -- Periodicals; Manufacturing processes -- Periodicals; 629.8
• DOI: 10.1177/0954405419889183
• Languages: English
• ISSNs: 0954-4054

Abstract

Burrs are undesirable materials beyond the work piece surface during drilling or other machining processes, thus this should be as less as possible during manufacturing. The experimental study has been conducted according to the full factorial design method. A total of 27 experiments were conducted by drilling on an Aluminum 6061T6 plate by choosing three factors and three levels of process parameters like drill diameter, point angle and spindle speed. In this research article, two predictive models, namely, adaptive neuro-fuzzy inference system and support vector regression, are developed using experimental data to estimate burr height and burr thickness. Then, these predictive models have been used to find out optimum process parameters for minimum burr height and burr thickness using genetic algorithm. It has been found that both the models are able to predict burr size and thickness with good accuracy, while the adaptive neuro-fuzzy inference system performs better than support vector regression.