Parametric optimization and experimental verification of 3D printer for enhanced mechanical properties
Loading...
Date
item.page.authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The Rapid prototyping techniques such as three-dimensional (3D) printing
newlinehave rapidly gained popularity in industry since material layers are added rather
newlinethan removed. Additive manufacturing creates objects from 3D CAD model data
newlineby layering materials, thus saving time and money. Fused deposition modeling
newline(FDM) is a 3D printing technique that prints thermoplastic, layer by layer. Various
newlineparameters affect the properties of the final printed object. The exact identification
newlineof variation in the properties of the printed object is still a very popular issue among
newlinethe researchers. In the present work, an effort has been made to identify the
newlineparametric influence on the printed object. The input parameters considered are
newlinelayer thickness, infill density, print speed, and extruder temperature. Four levels of
newlinethe four input variables have been identified. The combinations of input parameters
newlineduring the fabrication have been considered as per the Taguchi L16 Orthogonal
newlineArray. A total of 16 different samples have been fabricated, considering the
newlinerepetition rate of 3 per sample. The specimens of polylactic acid (PLA) have been
newlineprinted as per ASTM standards, using Fused Deposition Modelling (FDM). The
newlineproperties measured are the wear behavior, tensile strength, and flexural strength of
newlineprinted specimens.
newlineIn order to identify the parametric influence on wear, tensile strength and
newlineflexural strength, mathematical modeling has been done using regression and
newlineartificial neural networks (ANN). In the case of wear test, the results show that the
newlineaverage percentage variation in predicted and experimental values for regression
newlineand ANN models are, 5.04% and 1.94%, respectively. In case of flexural strength,
newlinevalue of R-square for regression is 86.92, while for tensile strength the value
newlineobtained is 86.08. Additionally, the percentage deviation between experimental
newlinevalues and model predicted values is calculated as per ANN model and the variation
newlineis 4.06 percent for flexural strength...