Please use this identifier to cite or link to this item:
|Title:||Vehicular load on bridge deck|
Loading and unloading
Hong Kong Polytechnic University -- Dissertations
|Publisher:||The Hong Kong Polytechnic University|
|Abstract:||In this thesis, two methods based on regularization technique are developed to identify the time-varying loads from vehicles moving on top of the bridge deck. The bridge deck is modeled either as a multi-span continuous beam when the bridge is narrow compared with its length, or as an orthotropic rectangular plate if otherwise. Computational simulations and laboratory tests are used to verify the feasibility and accuracy of these two methods. Two experimental setups are developed in the laboratory. One is for the beam model, and the other is for the orthotropic plate model. In the completion of the main objective to develop a moving force identification technique with a bridge-vehicle system, the following works have been performed: Firstly, the dynamic behavior of the continuous bridge deck under moving vehicles is analyzed. The influence of different parameters such as the road surface roughness of the bridge and the surface condition of the approach, multiple vehicles and their transverse positions, braking or acceleration on the bridge are studied using computational simulations and laboratory tests. Secondly, regularization on the ill-conditioned problem of indirect force identification is introduced to provide bounds to the identified forces. The Frequency and Time Domain Method (FTDM) (Law et al 1999) is selected in the study on the improvements due to regularization in both simulation and laboratory test results. The laboratory results from Time Domain Method (TDM) (Law et al, 1997) is also presented to compare the accuracy and the effectiveness of regularization in these two methods.|
Thirdly, two new methods based on regularization are proposed to overcome the deficiencies exhibited in existing methods. A new time domain method is developed to identify moving loads on a continuous beam from the measured structural vibration responses. This method gives exact solutions to the forces with improved formulation over existing methods for a more efficient computation. Another general method based on the finite element formulation is also developed to identify moving loads on a continuous beam. A generalized orthogonal function approach is proposed to obtain the derivatives of the bridge modal responses. The moving loads are identified using least squares method with regularization on the equation of motion in the time domain. This method is extended to identify the moving loads on non-uniform multi-span continuous Timoshenko beam. The comparative study between the results from using the Timoshenko beam theory and the Euler-Bernoulli beam theory is also included. Numerical examples on both single and multiple span bridges and the case of axle interaction forces from a four-DOFs vehicle on a triple-span bridge are used to demonstrate the feasibility and accuracy of these two methods, and factors affecting the errors in the identification are discussed. An experimental setup for the beam model is designed in the laboratory. The moving forces are identified from the measured strains using these two methods. The effect of non-uniform speed on the identified results when the forces are identified using a constant speed is also investigated. Fourthly, the bridge deck is simplified as a rectangular orthotropic plate and the two proposed methods are extended to identify the moving loads on the three-dimensional bridge deck. Computational simulations show the effectiveness and the validity of the proposed methods in identifying loads travelling along the central line or at an eccentric path on the bridge deck. An experimental setup for the bridge deck model is designed in the laboratory. The strains of the bridge deck are measured when the model car moves across the bridge deck along different paths and at different speeds. The moving loads on the bridge deck were identified from the measured strains using the two methods, and the reconstructed responses are calculated from the identified loads to verify the performances of these two methods.
|Description:||1 v. (various pagings) : ill. (some col.) ; 30 cm.|
PolyU Library Call No.: [THS] LG51 .H577P CSE 2001 Zhu
|Rights:||All rights reserved.|
|Appears in Collections:||CEE Theses|
PolyU Electronic Theses
Files in This Item:
|b15784769_link.htm||For PolyU Users||161 B||HTML||View/Open|
|b15784769_ir.pdf||For All Users (Non-printable)||14 MB||Adobe PDF||View/Open|
All items in the PolyU Institutional Repository are protected by copyright, with all rights reserved, unless otherwise indicated. No item in the PolyU IR may be reproduced for commercial or resale purposes.