FPA based optimization of drilling burr using regression analysis and ANN model. (February 2020)

• Record Type: Journal Article
• Authors: Mondal, Nripen; Mandal, Sudip; Mandal, Madhab Chandra
• Is Part Of: Measurement. Volume 152(2020)
• Publication Date: 2020-02
• Subjects: FPA -- Drilling burr -- Optimization -- ANN -- Regression -- Taguchi; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2019.107327
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: In this paper, a computational approach has been proposed to obtain optimum drilling process parameters. The second order Mathematical Regression model of burr size is developed using Minitab software. Regression model has been optimized using Flower Pollination Algorithm (FPA). Artificial Neural Network (ANN) analysis have been also performed to compare with the experimental and regression results. Abstract: The minimization of the burr formation in drilling process by selection of optimal process parameters is the aim of this research. The experimental study has been performed according to L27 orthogonal array for burr formation in aluminium alloy using HSS drill. The second order regression model of burr height was developed in Minitab16 from experimental data consist of process parameters i.e. spindle speed, feed rate, point angle and burr height. This model was optimized using Flower Pollination Algorithm, which helped to determine the optimal process parameters responsible for minimum burr height formation. The result shows that minimum burr height was formed for optimized values of speed 219 rpm, feed 0.04 mm/rev and point angle 113°. Next, ANN model has been developed to compare between experimental, regression and ANN model predicted burr height. For validation, new set of experiments were performed and validated with ANN and regression model with satisfactory accuracy.