Substructural identification of offshore structures using neural networks

A method is presented for estimating unknown parameters of offshore structures using a backpropagation neural network. Several techniques are employed to overcome the issues associated with a large number of degrees of freedom. They are the substructural identification and the submatrix scaling factor. The modal data such as natural frequencies and mode shapes are used as input to the neural network for effective element-level identification particularly for the case with incomplete measurement of modal data. A numerical example analysis on a jacket-type offshore structure is presented to illustrate the proposed procedure and to demonstrate the effectiveness of the method.