MU Digital Repository
Logo

Optimization of process parameters in drilling of glass fiber reinforced polymer composites using taguchi and system dynamics approach

Murthy, B R N (2014) Optimization of process parameters in drilling of glass fiber reinforced polymer composites using taguchi and system dynamics approach. Phd. Thesis thesis, Manipal Institute of Technology, Manipal.

[img] Archive (ZIP)
Abdul Khadar.zip - Submitted Version
Restricted to Registered users only

Download (5MB) | Request a copy

Abstract

Glass Fibre Reinforced Plastics (GFRP) composites find wide application in numerous manufacturing fields such as aerospace, automobile etc, because of their distinct properties such as low weight, high strength and stiffness. Even though the GFRP composite components are produced to near-net form, machining is usually required to meet the requirements associated with tolerances of assembly needs. Drilling is the most indispensable method for the fabrication of products with composite panels. The performance of these products is mainly dependant on surface quality and dimensional accuracy of the drilled hole. The quality of the hole drilled is influenced by the cutting conditions, composite properties, tool material and geometry. The material anisotropy resulting from fibre reinforcement considerably influences the quality of the drilled hole. Hence, precise machining must be performed to ensure the dimensional stability and interface quality. Since drilling in GFRP composites involves numerous variables which will influence the quality of the drilled hole, it is very essential to study the effect of these individual variables over the response. Control over the individual variables and optimization is important to obtain sensible results. The scope of the present research work is to experimentally identify the effect of drill point angle, drill diameter, material thickness, spindle speed and feed rate on thrust force, torque, surface roughness and delamination in drilling of composite material. The test specimen used is the glass fiber reinforced polymer (GFRP) composite laminate which was manufactured by hand lay-up method. The specimen consists of general purpose polyester resin as matrix, methyl ethyl ketone peroxide was used as the binder. The specimen was reinforced with E glass fiber. The weight faction of the composite is 44% which was confirmed by the burn test method. Design of experiment (DOE) method has been implemented to design the experiments. A full factorial design of five factors and three levels has been adopted in order to obtain the complete machining data. Hence, number of estimated runs are 35 = 243 and experiment has been done for a replication. Solid carbide drills of various diameter and drill point angles were used as the drilling tools. Systematic experiments have been designed and conducted in order to evaluate the machining characteristics during drilling. Since the quality of the drilled hole is influenced by various individual parameters, following individual variables are considered for the present work.  Material variables: thickness.  Machining variables: spindle speed and feed rate.  Cutting tool variables: drill diameter and drill point angle. The specimens are drilled without backing using a standard solid carbide drill bits of size 6mm, 8mm and 10mm. Drill point angles of 900, 1030, 1180 have been adopted to drill the specimen. In this work the drilling process was carried out with feed rates of 75, 110, 150 mm/min and spindle speed of 900, 1200, 1500 rpm. The machining has been done on TRIAC VMC CNC machining center and the machining condition is dry. The thrust force and torque developed during drilling were measured with the help of KISTLER Dynamometer and a charge amplifier, and the output data has been stored in computer. The surface roughness of the drilled hole was measured by Taylor Hobson Surtronic 3+ roughness measuring instrument. In order to measure the delamination the drilled holes were initially scanned using a scanner of 1200 dpi resolution and there after the dimensions of the scanned images are measured with the help of CATIA V19 software. Significance level of each parameter and the order of significance has been analyzed using ANOVA technique. To obtain the good quality holes, the optimization of the process parameters was done through Taguchi and Response Surface Methodology (RSM). The regression equations which show the correlation between individual variables and output responses were developed for each response parameter. The confirmation tests were conducted at optimum levels of process parameters and obtained results were compared with the predicted values to check the acceptance level of the regression model. The correlation of the thrust force with the delamination factor and surface roughness were also predicted. Further, an attempt has been made to develop the simulation through System Dynamics approach for the first time in the field of machining. In addition to the System Dynamics, simulations by ANN and RSM were also developed and comparison between the results of these three techniques has been made. To develop the System dynamics simulation the following steps are followed:  The mathematical equation has been developed with the experimental data which shows the relation between input and output parameters.  System Dynamics model (causal loop diagram) has been developed using VENSIM® software and the mathematical equation was entered in the model.  The model was run for various combinations of parameters and the corresponding responses were tabulated.  With the help of tabulated results, simulation graphs were plotted using MATLAB® software. The ANN simulation was done using genetic algorithm multilayer perceptron ANN model which consists of single hidden layer. Out of 243 data, 170 data (70%) were used for training and 73 (30%) data were used for testing the neurons. The number of epochs used are 2000 and number of iterations considered are 100. After trial and error, the hidden neuron number is set to 58. The learning rate of the model is 0.0158 and the momentum rate is 0.0033.

Item Type: Thesis (Phd. Thesis)
Subjects: Engineering > MIT Manipal > Humanities and Management
Depositing User: MIT Library
Date Deposited: 28 Mar 2016 11:12
Last Modified: 28 Mar 2016 11:12
URI: http://eprints.manipal.edu/id/eprint/145672

Actions (login required)

View Item View Item