Study On Identification Methods Of Modal Parameters And Flutter Derivatives Of Full-bridge Aeroelastic Model

This dissertation summaries the development of the theory of bridge wind engineering, introduces wind tunnel test which is still the most effective and reliable way to study wind-induced vibrations of long-span bridges, and modal test of bridge aeroelastic model in wind tunnel tests plays a pivot role. The dissertation studies time domain modal parameters identification methods of bridge aeroelastic model, which are suitable for wind tunnel tests (only when output response data without loading the input data). The methods include: Ibrahim Time Domain (ITD) method, Eigensystem Realization Algorithm (ERA) method and Stochastic Subspace Identification (SSI). Based on the double modal parameters of full-bridge aeroelastic model, the feasibility analysis and further study on flutter derivatives identification of its girder have been discussed. The main contents and results of this paper are as follows:1. The development of bridge wind engineering theory and wind tunnel test techniques have first been reviewed, and the importance of modal parameters of wind-induced vibration of bridges and a brief introduction about main methods of existing modal parameters identification have been introduced.2. A comprehensive explanation of precision, work principle and environment of non-contact3D displacement instrument are introduced. This paper studies the advantages and key operation points of the instrument which is used in aeroelastic model for modal identification tests. In-depth research has been taken for preprocessing methods of the data which is obtained from the instrument, the research includes:digital filter design, non-stationary response of the principle of Empirical Mode Decomposition (EMD) and reconstruction and reduction effect of random decrement technique.3. Time domain modal identification algorithms which are suitable for the aeroelastic model in wind tunnel are focused on, the algorithms include:ITD method, ERA method and SSI method. The three algorithms has improved to improve the computational efficiency and accuracy, and the MATLAB(?) programming language is used to achieve the results of automatic modal parameters identification. A simple numerical example test is used to prove the correctness and reliability of the algorithm. Further researches of the different algorithms have been discussed, which includes:anti-noise capability, error analysis, application range and algorithm defects. As for the instability of damping recognition, this paper uses a method which combines of stability and hierarchical clustering.4. Modal identification test of Maputo full bridge aeroelastic model is used as engineering background. Detailed modal identification procedure has been discussed. Modal identification algorithm (ITD, ERA and SSI) are used in automatic modal parameters identification, the results, such as frequencies, mode shapes and damping coefficients of the bridge, are obtained and then they are compared with the theoretical value. Finally, error source of modal parameters identification of full-bridge aeroelastic model is analyzed.5. The application of modal parameters identification method in the full bridge aeroelastic model to identify flutter derivatives has been studied. An ideal flat section which acts as an example is used for simulation analysis of the free decay response and random wind loads response, the simulation analysis suggests that the method using modal parameters identification methods (ITD, ERA and SSI) to identify flutter derivatives of main beam is feasible.