Research On Identification Method For Flutter Derivatives Of Bridges Decks And Realization By MATLAB

The identification of flutter derivatives is the foundation and hot issues of flutter research of long span bridges, this paper reviewed the development course of theory of bridge flutter and flutter derivatives, methods for identification of flutter derivatives of bridge decks were summarized. Based on the section model of two-dimensional bending-torsional coupling free vibration of wind tunnel test, the modal parameter identification method, data pre-processing, data interception, and several other aspects were studied and discussed in this paper. A set of efficient and accurate method for identification of flutter derivatives and the matching MATLAB calculation program were given. The main research contents of this paper are as follows:(1) On the establishment of relationship between flutter derivatives and the free vibration modal parameters, three kinds of modal parameter identification method(the ITD method, the MITD method optimized by this paper and the LSCE method) were introduced. And a single degree of freedom simulation test were curried out to practically analysis and compare the characteristics of these methods and their inapplicability of directly applied to the identification of flutter derivatives.(2) Based on the review of existing least square iterative method(ULS method and MLS method), the method of weighted least squares iterative(WLS method) is derived by introducing a weighted matrix which is consistent with the characteristic of the test data. The simulation experiment of the two degree of freedom model is carried out, and the WLS method is proved to have better anti-noise ability and computational efficiency by immunity testing and initial value sensitivity testing.(3) Time-history curve envelope method was given to provide the initial values for least square iterative method. In order to solve the problem that the difference between the actual collected data in experiment and the effective data, a practical method based on random sample consensus algorithm(RANSAC) was proposed and verified through an example.(4) For the weak signal, large noise, decaying fast characteristic of free response data that obtained through the initial incentive in high wind speeds, using wind load as ambient excitation, the random decrement technique was applied in wind tunnel test of section model, and the free attenuation signal of system was obtained successfully.(5) Based on MATLAB language, the calculation modules corresponding to several previously proposed methods were compiled and a complete set of visual program for identification of flutter derivatives was developed. And this program was used in section model wind tunnel test to improve the accuracy and efficiency of the identification of flutter derivatives.