The Research Of Frame Structure Feature Information Extracting And Physical Parameters Identification

The reliability of the important engineering structure has important influence on the society and economy. Assessing the structure exactly is the basic premise of the structure work reliable. The structure physical parameters are important parameters of the civil engineering structure. It has a close relationship with the performance of the engineering structure, a more important status in the safety examination of the structure and an extremely vital significance to identify these parameters. So it needs a further research on the methods of identifying the structure physical parameters. Currently the research that the structure physical parameters identification is mainly combines with some practical engineering, shear structure is a most popular used of structure in the building, ambient ground motion excitation has already become an ideally excitation resource because of its advantages such as easy to realize, low cost and little impacting on environment etc. This paper devoted in the study mainly in the methods of physical parameter identification according to above-mentioned practical engineering background.This paper can be divided into six parts.In the first part (chapter 1), it is an introduction. The research background and present research condition are reviewed and summarized.In the second part (chapter 2), it is a basic theory part. It introduces the mathematic model suitable for physical parameters identification; basic identification principle and least squares algorithm in structural physical parameters inversion identification are also introduced.In the third part (chapter 3), it mainly devoted in study of the noise influence on the results with the structure physical parameters inversion identification methods. According to the noise frequency, some research is devoted in the influence of results with random white noise and low frequency noise by numerical imitate. At the same time, considering of the characteristic of linear dynamic system, put forward a new denoising method which it extracts the structure feature information from the signal, this method can eliminate noise effectively. Numerical example shows that the stiffness identification results are more accurate by identifying the structure feature information.In the fourth part (chapter 4), it studies the complete information structure and the sub-structure physical parameter inversion identification methods by using a shear structure model, and presents a numerical integration method of speed and displacement, and makes a discussion on the problems of initial integration and integral error etc. Finally it identifies a six-floor shear structure example.In the fifth part (chapter 5), it is an experimental research on steel-concrete frame model physical parameters identification. By using a six-floor steel-concrete frame model, acquires a more accurate parameter through static experiment, then acquires each layer's stiffness by D-value method and experiment "modifying" idea, According to measured acceleration information, identifies the layer-stiffness with the complete information structure physical parameter inversion identification method and the sub-structure physical parameter inversion identification methods and verifies that the methods are effective in engineering structure dynamic damage detection..In the sixth part (conclusion), it summarizes the main content of this paper and predicts the future development of this field.