[雙語翻譯]--橋梁外文翻譯--對斜鐵路鋼橋進行高等的動態(tài)有限元分析（原文）

1、Engineering Structures 33 (2011) 181–190Contents lists available at ScienceDirectEngineering Structuresjournal homepage: www.elsevier.com/locate/engstructAdvanced dynamic finite element analysis of a skew steel railway b

2、ridgeG. Kaliyaperumal ?, B. Imam, T. RighiniotisFaculty of Engineering and Physical Sciences, University of Surrey, Guildford, GU2 7XH, UKa r t i c l e i n f oArticle history: Received 24 March 2010 Received in revised f

3、orm 6 August 2010 Accepted 4 October 2010 Available online 3 November 2010Keywords: Steel railway bridge Dynamic analysis Field measurements Frequency Stressesa b s t r a c tThe aim of this paper is to present advanced m

4、odelling techniques for dynamic analysis of steel railway bridges. Finite element analyses of a case study skew bridge are carried out and the results are compared with available field measurements. Initially, eigenvalue

5、 analyses of different models are carried out in order to obtain the fundamental mode shapes and bridge frequencies and to assess the capability of each model to capture the dynamic behaviour of the bridge. Single-span,

6、three-span and full bridge models are investigated with different elements such as shell, beam and combinations of these. A very good agreement between the fundamental dynamic properties of the bridge and empirical resul

7、ts is found. Following the eigenvalue analyses, time history dynamic analyses are carried out using the full bridge model. The analyses are carried out for different train speeds and the strain histories are compared wit

8、h available field measurements. In terms of fatigue assessment, the mean stress range values obtained from the strain histories at selected locations on the bridge members are also compared to each other. The results sho

9、w that a full bridge model using a combination of beam and shell elements is a reasonably accurate and computationally efficient way of capturing the dynamic behaviour of a bridge and estimating the mean stress range for

10、 fatigue damage calculations. © 2010 Elsevier Ltd. All rights reserved.1. IntroductionDynamic analysis of railway bridges has received considerableattention during the last few decades. This can be attributed to the

11、 considerable development the finite element (FE) method and the increase in the capabilities of computers. Static analyses of bridges can be easily carried out using the FE method. However, dynamic analyses are much mor

12、e demanding and, therefore, optimum models which are capable of capturing dynamic behaviour reasonably well and with time efficiency need to be developed.Dynamic effects on a bridge can have an effect on the ultimatelimi

13、t state behaviour of the bridge by amplifying the maximum stresses experienced by different members. During design, dy- namic amplification factors are usually employed in order to take into account dynamic effects and a

14、mplify the statically-calculated stresses. In many cases, these factors may be over-conservative. On the other hand, dynamic effects may also have an impact on the fa- tigue behaviour of a member or connection by increas

15、ing the stress ranges and number of stress cycles experienced by them. Since fa- tigue behaviour is very sensitive to applied stresses, an accurate estimation of the dynamic stresses in a bridge is fundamental to- wards

16、its fatigue assessment.A number of past studies have attempted to quantify thedynamic impact caused by trains on railway bridges [1–3] and? Corresponding author. Tel.: +44 07990636456; fax: +44 1483682135.E-mail address:

17、 g.kaliyaperumal@surrey.ac.uk (G. Kaliyaperumal).motor vehicles on highway bridges [4–6]. The dynamic effect induced by the moving vehicle is considered by multiplying the static load with the Dynamic Load Factor (DLF),

18、where DLF = 1 + Dynamic Load Allowance (DLA). The DLF has to be evaluated by a probabilistic approach rather than considering it as a deterministic value. The importance of the factors affecting the DLA for highway bridg

19、es has been highlighted and an equation for calculating the DLA has been presented in [4–6]. In [2], the dynamic stresses were used to estimate the remaining fatigue life of different elements on a railway truss bridge.

20、A number of other investigators have attempted to calibrate their finite element model of the bridge with field measurements [7–9]. In [7], eigenvalue analysis of a truss bridge was carried out to determine its fundament

21、al mode shapes, and the results obtained from dynamic analyses under the passage of a locomotive were compared with field measurements. The calibrated model was then used to assess the capacity of the bridge.The majority

22、 of the dynamic studies on railway bridges haveemployed finite element models using beam/frame elements which may be sufficient for the prediction of the overall behaviour of bridges. However, beam/frame elements are not

23、 capable of cap- turing local connection behaviour and out-of-plane movements and distortions of girders which may be critical in the case of plate girder bridges. The out-of-plane distortions may induce secondary stress

24、es in welded plate girder skew bridges, which may lead to fatigue damage [10–12].The aim of this paper is investigate the effects of differentmodelling assumptions on the dynamic behaviour of steel bridges0141-0296/$ – s

25、ee front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.engstruct.2010.10.003G. Kaliyaperumal et al. / Engineering Structures 33 (2011) 181–190 183(a) Plan at mid-span. (b) Cross-section at location

26、C.Fig. 3. Details of strain gauge locations at bridge span 7–8.Fig. 4. Typical measured strain (at point S2) for different locomotive speeds.Fig. 5. Single-span shell element FE model of the bridge with bracings.the brid

27、ge. The effect of bracings was also investigated via the single-span FE model by developing a shell model with and a model without bracings. Fig. 5 shows the shell-element model of the single span with the bracings inclu

28、ded. In all the shell-element models, the stiffeners in the main girders were also modelled. In terms of the three-span bridge models, one model was developed using shell elements for all spans, as shown in Fig. 6 wherea

29、s the other model was developed by a combination of shell and beam elements. The latter model is shown in Fig. 7 where it can be seen that shell elements were employed for span 7–8 whereas beam elements were used to mode

30、l the adjacent two spans. For the full bridge model, which is shown in Fig. 8, shell elements were used only for span 7–8, the remaining spans being modelled with beam elements. In both the three-span and six-span FE mod

31、els, the intermediate supports were modelled as simply supported.Fig. 6. Three-span shell element FE model of the bridge.Fig. 7. Three-span beam and shell element FE model of the bridge.Fig. 8. Beam and shell element FE