Artificial Neural Networks for Machining


Bayraktar Ş. , Alparslan C.

in: Advances in Sustainable Machining and Manufacturing Processes, Kishor Kumar Gajrani,Arbind Prasad,Ashwani Kumar, Editor, CRC, New York , New York, pp.189-204, 2022

  • Publication Type: Book Chapter / Chapter Research Book
  • Publication Date: 2022
  • Publisher: CRC, New York 
  • City: New York
  • Page Numbers: pp.189-204
  • Editors: Kishor Kumar Gajrani,Arbind Prasad,Ashwani Kumar, Editor

Abstract

Artificial neural networks (ANNs) are computer systems developed with the aim of automatically realizing abilities such as creating and discovering new information by taking an example from the human brain without any assistance. It emerged as a result of mathematical modeling of the learning process. ANNs are used for prediction, classification, data association, clustering, filtering, interpretation, optimization, and control thanks to their nonlinear structures and continuities. In addition, it is used in many fields such as image processing (face, motion, and object recognition), automotive, aviation, and manufacturing industries, as well as natural language processing (voice assistant, emotion analysis) computational biology (DNA sequencing, tumor detection, drug discovery). Because ANNs can process highly nonlinear situations, many parameters and incomplete information. An ANN structure consists of inputs, weights, addition function, activation function, and outputs. Learning abilities improve thanks to the parameters entered for each operation. Experimental outputs, such as surface roughness, cutting force, and tool wear, can be successfully predicted according to independent variables for each machining technique in manufacturing. ANN models used in different machining techniques according to independent variables are presented in this chapter. In addition, the relationships between the current studies in the literature and the mathematical models obtained in these studies and the experimental outputs are discussed.