Methods And Research Of Inverse Problems About Structural Parameters Of Dynamic Behaviors

In this paper many a method has been discussed how to deal with the inverse problem of structural parameters of dynamic behaviors, such as theoretical method, experimental method, theoretical conjunct with experimental method and computer simulation. Some of them are summed of others' conclusions; some are established by author himself.Firstly, by inducing the formula we proved that it is possible to identify the structural parameters of bar and beam input one dimension wave, and the solution exists under the method of normalization.then with using the related structural response data when earthquake, combined with Genetic Algorithm (GA), a new method named general statistics mean method based on GA is founded, pointed out that input, output and the parameters of modal have a one-to-one relationship for a linear structure loaded by non-monotonic forces. By introducing global optimal algorithm such as GA, we effectively solved the nonlinear inverse problem of structural parameters. According to the inverse problem theory of modal, physical parameters are calculated by measured modal data. After considering of analysis of Finite Element method and GA, the fundamental parameters, for example, stiffness, dampness etc., are identified. Advanced discuss of the location of detects in structures are forward using residual forces caused by structural changes.At last, with use of neural network capable of non-linear, self-learned, error-capability and redundancy, structural damage identification by samples' learning, testing and applying is practiced. Detailed research on how to select the form of network, parameters and the components of samples also proceeds.The results of reinforced concrete frames from tests show that the methods set up in the paper are acceptable.