Artificial neural networks applied to prediction of surface roughness in dry drilling of some polymers
Author:
- V TabacaruVTV Tabacaru
Open Access
- 1 September 2020
- conference paper
- Published by IOP Publishing in IOP Conference Series: Materials Science and Engineering
- Vol. 916 (1)
Abstract
Polymers become more and more attractive for automotive and aerospace industries due to their remarkable mechanical, thermal and electrical properties that make these materials suitable for many industrial applications. Machining of polymers is of a great interest among researchers and engineers due to the possibility of replacing expensive materials with plastics that have similar mechanical characteristics, but of a lower cost. Drilling is the most common mechanical machining operation in manufacturing of parts. Within this context, it is essential to analyze this cutting process and find the best solution for controlling the output parameters of the process, such as surface roughness and cutting forces. The present work concerns the developing of an artificial neural model for the prediction of surface roughness in dry drilling of some polymeric materials: high density polyethylene (grade HDPE 1000), polyamide (grade PA6) and polyacetale (grade POM-C). The neural model was built based on trial-and-error method, by modifying the number of hidden layers and the number of hidden neurons on each layer. The experimental plan was designed to be suitable for artificial neural network prediction. The aim of experimental work was to study the effect of cutting parameters (spindle speed, feed rate and drill diameter) on the quality of machined surface (surface roughness). This paper proposes a neural model that is able to predict the surface roughness considering not only the cutting parameters, but the type of material. Moreover, a study concerning the accuracy of the neural model depending on the number of hidden layers and the number of hidden neurons on each hidden layer was carried out. The predicted values of roughness were compared with experimental data in order to determine the precision of the neural model.Keywords
This publication has 16 references indexed in Scilit:
- The complementary graphical method used for profiling side mill for generation of helical surfaceIOP Conference Series: Materials Science and Engineering, 2017
- Modeling and optimization of turning process parameters during the cutting of polymer (POM C) based on RSM, ANN, and DF methodsThe International Journal of Advanced Manufacturing Technology, 2017
- A Review Paper on Effects of Drilling on Glass Fiber Reinforced PlasticProcedia Technology, 2014
- Optimization of process parameters in drilling of GFRP composite using Taguchi methodJournal of Materials Research and Technology, 2014
- Regression and ANN models for estimating minimum value of machining performanceApplied Mathematical Modelling, 2012
- A review of mechanical drilling for composite laminatesComposite Structures, 2012
- Surface roughness analysis in high-speed drilling of unreinforced and reinforced polyamidesJournal of Composite Materials, 2011
- Application of Taguchi methods in the optimization of cutting parameters for surface finish and hole diameter accuracy in dry drilling processesThe International Journal of Advanced Manufacturing Technology, 2008
- Polymer Property Prediction and Optimization using Neural NetworksIEEE Transactions on Neural Networks, 2006
- Artificial neural networks applied to polymer composites: a reviewComposites Science and Technology, 2003