Moving Force Identification Of Bridge Based On ITDM And The Application Of DOFS In Moving Force Identification

As the development of our country's communication cause, along with the environmental sustainability requirements, more and more bridges are needed to be constructed. And the moving force indirect identification through the dynamic responses of bridge under load cases can be a better approach for the research of bridge fatigue failure mechanism, overloading control of vehicles, and the improvement of bridge design theory. But to put this technology into effect, there is still much work needed to be done. The two primary problems are measuring technology advancement and identification model improvements, involving many disciplines. In this paper, the distributed optic fiber sensor technology, ITDM (improved time domain method), and the mode selection methods are introduced, and the following works are focused:Firstly, the feasibility of DOFS (distributed optic fiber sensor) used in the moving force identification of bridge is analyzed. The problems needed to be solved when introducing DOFS into load identification are theoretically analyzed. For the DOFS has many advantages in measurement, such as well sensibility, well continuity and well noise immunity, it can be used to achieve real-time, complete and reliable signals. And the products gained in the research of DOFS used in other field provide many valuable references for its use in load identification. But these are far from sufficient, which need the close collaborate of both structural research and measurement-control technology research.Secondly, the traditional truncated mode in modal analysis may arouse result overflow for the numerical sickness, so the mode-selection method is introduced. This method is set up based on the concept of signal-to-noise ratio and a rule and formula is developed. And the practical example shows that this method can be used to select mode in practice.Thirdly, a bridge model is set up to study the engineering use of ITDM. Based on the conception of RPE (relative percentage error), the influences of BFN (basis function number), MN (mode number), MPN (measuring point number), measuring point position, and processing time on the load identification are studied qualitatively and quantitatively. And a conclusion is drawn that more BFN can improve identification precision but also multiply the operating time, and the identification precision of the moments when the vehicle gets in and off the bridge deteriorates as the BFN multiply. The conclusion also shows that as the MN grows the precision improves at first then decreases, the more the MPN the better the precision, and the identification accuracy can be improved by increasing MPN without additional time.