Mathematically Modeling and Optimization by Artificial Neural Network of Surface Roughness in CNC Milling

Mathematically Modeling and Optimization by Artificial Neural Network of Surface Roughness in CNC Milling – A Case Study

Author Bekir Cirak
DOI

Country : Turkey
Subject : Mechanical Engineering

Keywords : Artificial neural networks; Modelling; Material processing; Surface roughness; Prediction

Abstract

In this study, steel AISI 1050 is subjected to process of face milling in CNC milling machine and such parameters as cutting speed, feed rate, flow rate of coolant, depth of cut influencing the surface roughness are investigated. Experimentally, an artificial neural network (ANN) approach for modeling of surface roughness in AISI 1050 steel material. These data have been presented to train a multi layered, single directed, hierarchically connected ANNs using Scaled Conjugate Gradient (SCG) and Levenberg-Marquardt (LM) Back Propagation algorithms with the logistic sigmoid transfer function. The outcomes demonstrated that, the ANN based model have been very successful and the testing data produced of errors.The surface roughness (Ra) is chosen as a measure of surface quality. The data set from major experiment is employed for training a feed forward three layer backpropagation ANN. The developed ANN model is tested on the other combinations of the cutting parameters in the given ranges, which are not included in the training process. The results of calculations are in good agreement with the experimental data confirming the effectiveness of ANN approach in modeling of surface roughness in CNN milling. The ANN model uses a multi-layer feed forward network architecture and was trained with experimental data using backpropagation. The ANN has 4 input neurons, 7 neurons in the hidden layer and 1 output neurons. The ANN model can predict the experimental results quite well with correlation coefficients in the range of 0.95 to 0.98.

Download