With the increasing size and complexity of machines and vessels, the easy health monitoring method is becoming necessary. In this paper the possibility of using a multilayer perceptron network trained with the backpropagation algorithm for detecting location and size of the fault in structural element is studied. The finite element model of the structure considered is a slender free-free beam, using an integrated software package, ANSYS, to estimate the changes in various modal parameters, caused by a notch and a damage modeled fault. The basic idea is to train a neural network with simulated patterns of the relative changes in natural frequencies and corresponding sizes (or shapes) and location of faults in order to recognize the behaviour of the damaged as well as the undamaged structure. Subjecting this neural network to measured values should imply information about the fault sizes and locations. The training data are obtaining by the values using the finite element method. Additionally, a series of test were carried out using the response measurement equipment to experimentally determine the effect of the imposed defect. The results show that a neural network trained with simulated data is capable for detecting location of a fault in a beam when the network is subjected to experimental data.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £55 +P&P)