Artificial Neural Network Based on Genetic Learning for Machining of Composite Materials

• Type: Article in International Conference Proceedings Book
• Date: 2006

Authors

Carlos Alberto Conceição António (Author) FEUP; Paulo Davim (Author) FEUP; Vitor Lapa (Author) Other

Conference proceedings International

Title: Mechanics and materials in design

Pages: 1-10

5th International Conference on Mechanics and Materials in Design (M2D'2006)

Porto, Portugal, July 24-26, 2006

Scientific classification

CORDIS: Technological sciences

Other information

Abstract (EN): In this paper an Artificial Neural Network (ANN) aiming the efficient modeling of a set of machining conditions in the orthogonal cutting of composite materials is presented. The experimental procedure considers process parameters as cutting speed and feed rate, the type of insert of the tool and the type of workpiece material in order to obtain a set of results used for ANN learning. The supervised learning of the ANN is based on a genetic algorithm with an elitist strategy. Input, hidden and output layers model the topology of the ANN. The weights of the synapses, the bias for the hidden and output nodes and the number of neural nodes of the hidden layer are used as design variables. Sigmoid activation functions are used in hidden and output layers. The square error between experimental and numerical results is used to monitoring the learning process aiming to obtain the completeness of modeling of the machining process.

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 10