Damage localization from vibration data using hierarchical a priori assumptions

An inverse method for damage identification is described. The method assumes that a forward model is available to predict hypothetical vibration data, for an arbitrarily varying profile of the local stiffness of the structure. In general, the inverse problem of identifying this stiffness profile from the observed vibration data is an ill-posed problem, therefore some form of regularization is required. In the case of this method, the regularization results from a Bayesian formulation of the a priori assumption that the stiffness change is smooth almost every where, but could have an unknown number of discontinuities at unknown locations.