Research Of Damage Indentification Of Building Structures

Frequent occurrences of natural corrosion and other lots of complicated functions cause damage in structures inevitably under an explicated operated environment. The correct evaluation, detection and repair of structural damage play an important role in greatly reducing the loss of lives and properties . The thesis explicates the theories and methods of evaluations on the building structures , meanwhile , systemically study inherent reasons based on the modal parameters and artificial neural network. The research is mainly consisting of the followings:(1) analyze the reasons of damage of structures and elaborate the theories and methods of evaluations on the building structures. Reinforced concrete three level damage model are also discussed from the basic conception of damage mechanics.(2) based on the structural dynamics , systemically study inherent reasons based on the modal parameters. Prevalent testing method lead from the structural dynamic parameters are also analyze and summarized.(3) with the strong non-linear mapping ability of the artificial neutral network, an ameliorated program of BP neutral network was worked out, we combine modal analysis method with neural network technique , taking modal parameters to construct identification parameters diagnostic vector as input vector to solve identification problems step by step.(4) the damage of the building structures will cause the change of the structural d ynamic p arameters, the r elationship b etween frequency, mode, mode variance and damage location together with damage degree are also analyze.(5) during course of practical damage identification, actual error inevitably effect on the result. study of the relationship of the measure error impact on the result of the artificial neutral network , and some suggestions was proposed on the raise the identified accuracy of the artificial neutral network.