Resumo

BARRERA C., Gerardo; CARREON G., Guillermo  e  RUIZ M., Alberto. Quality Control of Ceramic Pieces by Means of Vibration Analysis and Artificial Neuronal Networks. Comp. y Sist. [online]. 2005, vol.8, n.3, pp.187-195. ISSN 2007-9737.

The main purpose of this work is to find a fast and reliable diagnostic technique, for detection of internal or external cracks on ceramic specimens. This technique uses artificial neuronal networks (ANN) as a classification tool. The proposed methodology is mainly oriented to ceramic mosaics. A piezoelectric type accelerometer was used to detect the mechanical vibration and convert it to an electrical signal. The detected signal was digitized using a DAQ card. All the stages: data acquisition, signal conditioning, signal processing and neuronal networks training were carried out using G program language routines. The designed system runs under Windows® 2000 platform and virtual instrumentation. To do the classification process the proposed algorithms are Kohonen and Fuzzy C Means.

Palavras-chave : Quality control; ceramics; transducers; A/D signal conversion; Fourier rapid transform; artificial neuronal networks; virtual instrumentation.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )