Bayesian Fusion Theory and Its Application to the Identification of Structural Damages

A Bayesian fusion theory (BFT) method to localize multiple structural damage by use of modal strain energy and frequency data is presented. First, modal strain energy dissipation ratio theory is applied to obtain local identification information and the acquired data is converted into probability values; then, a frequency change damage detection method is introduced to obtain complementary identification information and the acquired data is also converted into probability values. Finally, the Bayesian fusion theory is proposed to integrate all probability values. Thus, the global identification information is acquired, which is more accurate than the local information. A truss structure is analyzed as a numerical example to illustrate the performance of the proposed method. The simulation results demonstrate the proposed BFT method can perfectly identify damage sites, and the identification results of the BFT method are better than those of both the modal strain energy dissipation ratio method and frequency change damage detection method.