In this paper, the determination of defects in concrete structures using an ultrasound technique is discussed. A diagnostic model for concrete pillars by means of Multi Layer Perceptron neural networks is developed to locate and classify the defects. Finite Elements numerical techniques have been used to model a concrete pillar of specified size (i.e., rectangular cross section and 2 meters in span) affected by defects of different position and size. The numerical analyses enable to obtain several received signals containing the fault information. These signals have been processed by a feature extractor system, whose purpose is to reduce the data dimensionality and to compute suitable features. Results showed good accuracy in the identification of the position and entity of the faults.
Artificial Neural Networks for the Non Destructive Testing of Concrete Structures
CANNAS, BARBARA;CARCANGIU, SARA;CAU, FRANCESCA;FANNI, ALESSANDRA;MONTISCI, AUGUSTO;TESTONI, PIETRO
2007-01-01
Abstract
In this paper, the determination of defects in concrete structures using an ultrasound technique is discussed. A diagnostic model for concrete pillars by means of Multi Layer Perceptron neural networks is developed to locate and classify the defects. Finite Elements numerical techniques have been used to model a concrete pillar of specified size (i.e., rectangular cross section and 2 meters in span) affected by defects of different position and size. The numerical analyses enable to obtain several received signals containing the fault information. These signals have been processed by a feature extractor system, whose purpose is to reduce the data dimensionality and to compute suitable features. Results showed good accuracy in the identification of the position and entity of the faults.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
