Comparison of response surface model with neural network in determining the surface quality of moulded parts

In this study, response surface (RS) model and an artificial neural network (ANN) are developed to predict surface roughness values error on mold surfaces. In the development of predictive models, cutting parameters of feed, cutting speed, axial–radial depth of cut, and machining tolerance are consi...

Full description

Saved in:
Bibliographic Details
Published in:Materials in engineering 2007, Vol.28 (2), p.459-465
Main Authors: Erzurumlu, Tuncay, Oktem, Hasan
Format: Article
Language:English
Subjects:
Artificial neural network
Cutting parameters
Milling
Mold surfaces
Response surface model
Surface roughness
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this study, response surface (RS) model and an artificial neural network (ANN) are developed to predict surface roughness values error on mold surfaces. In the development of predictive models, cutting parameters of feed, cutting speed, axial–radial depth of cut, and machining tolerance are considered as model variables. For this purpose, a number of machining experiments based on statistical three-level full factorial design of experiments method are carried out in order to collect surface roughness values. An effective fourth order RS model is developed utilizing experimental measurements in the mold cavity. A feed forward neural network based on back-propagation is a multilayered architecture made up of one or more hidden layers (2 layers–42 neurons) placed between the input (1 layer–5 neurons) and output (1 layer-1 neuron) layers. The response surface model and an artificial neural network are compared with manufacturing problems such as computational cost, cutting forces, tool life, dimensional accuracy, etc.
ISSN:0261-3069
DOI:10.1016/j.matdes.2005.09.004