[雙語翻譯]--外文翻譯--小跨度預制混泥土拱橋掩埋時的性能評估（原文）

1、Performance Assessment of a Precast-Concrete, Buried, Small Arch BridgeManoochehr Zoghi1 and Daniel N. Farhey2Abstract: Experimental field load-test and finite-element analysis were carried out for the performance assess

2、ment of a precast-concrete, modular, three-sided, low-profile, buried, arch bridge system. Finite-element analysis incorporated soil modeling and soil–structure interaction at service and limit load levels. The analytica

3、l study simulates step-by-step incremental phases of construction and service loads. The finite-element model was calibrated based on the experimental field assessment, to provide a better correlation between the analyti

4、cally predicted behavior and the actual response of the structure. The study validates the incorporation of various soil models and soil–structure interaction characteristics, to allow a more cost-effective bridge design

5、.DOI: 10.1061/?ASCE?0887-3828?2006?20:3?244?CE Database subject headings: Experimentation; Finite element method; Structural analysis; Soil-structure interaction; Concrete, precast; Buried structures; Bridges, arch.Intro

6、ductionIdentification of load and failure mechanisms through finite- element modeling and analysis, integrated with experimental field testing, offers promise for the practical evaluation of structural performance and in

7、tegrity of certain forms of bridges. Since boundary conditions greatly affect modeled structural behavior, a thorough assessment of a buried structure requires consideration of the geomechanical conditions. Finite-elemen

8、t analysis has en- abled soil modeling and the incorporation of soil–structure inter- action in design. However, the reliability of the analysis is highly dependent on the accuracy of the soil models and properties that

9、are used for the given site conditions. In planning a bridge replacement, one of the primary concerns is economical construction. Precast-concrete modular units are often a preferable solution for small bridge replacemen

10、ts due to their low initial cost, rapid installation, and low maintenance. In order to approve the use of longer span modular bridge units, departments of transportation require field performance assess- ment. Accordingl

11、y, a newly constructed precast-concrete replace- ment bridge in Miami Township, Montgomery County, Ohio ?MIA-033-0.54, on Gay Road, intersection of Crains Run Road? was considered as a test specimen for the experimental

12、phase of this research ?Zoghi 1994; Zoghi and Hastings 2000?. An experimental field load test and a finite-element analysiswere carried out on this bridge, to investigate the load carrying capacity and assess the limit p

13、erformance per the AASHTO Guide ?AASHTO 1989a? and AASHTO Manual ?AASHTO 1994?. To determine and evaluate the most appropriate soil model, the finite-element model of the bridge was calibrated by comparing the analytical

14、 results with the field-measured re- sponses. Determination of the most appropriate soil model and its limitations may allow the design of more efficient and, therefore, more cost-effective bridge structures.Objectives a

15、nd ScopeThe present investigation intended to: ?1? determine the integrity of the bridge superstructure and its components; ?2? determine the load–deformation character; and ?3? identify the ultimate load- carrying capac

16、ity incorporating soil–structure interaction. The objective of this paper is to describe the application of a limit performance assessment of a precast-concrete, modular, three-sided, buried, arch bridge system. This app

17、lication enabled a feasible evaluation of the load-carrying capacity and soil– structure interaction of the buried structure, under the actual site conditions. The present study demonstrates the use of condition- assessm

18、ent techniques in conjunction with conventional engineer- ing practice. Integration of field experimental technologies with comprehensive finite-element modeling and analysis enabled an objective evaluation of the system

19、’s behavior. The scope of the present application consisted of: ?1? finite- element modeling and analysis of the bridge system, including the bridge structure, soil, and soil–structure interaction; ?2? field test- ing un

20、der controlled and gradually increasing static load to moni- tor the relevant stresses and displacements at service and limit load levels; ?3? analytical simulation for the evaluation of various soil properties and soil

21、models; ?4? field calibration of the finite- element model; ?5? material sampling and testing to quantify the material properties used in the analysis; and ?6? serviceability and limit performance assessment of the bridg

22、e system.1Associate Professor, Civilapproved on July 28, 2005. This paper is part of the Journal of Performance of Constructed Facilities, Vol. 20, No. 3, August 1, 2006. ©ASCE, ISSN 0887-3828/2006/3-244– 252/$25.00

23、.244 / JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES © ASCE / AUGUST 2006ModelingOne of the study goals was to compare the finite-element analysis results to those obtained from the experimental load test, and ac

24、cordingly calibrate the former. Consequently, a finite-element model of the bridge was developed. This preliminary model was analyzed to characterize the bridge and identify its critical re- sponses. The modified finite-

25、element program CANDE, devel- oped by the Federal Highway Administration ?FHWA? ?Katona et al. 1976; Musser 1989?, was employed for the analytical as- sessment. This program enables comprehensive soil modeling and soil–s

26、tructure interaction study of buried structural systems, be- yond the conventional elastic range into the plastic state. The two-dimensional finite-element analysis was based on plane- strain methodology, whereby the mod

27、el is considered infinite in length and with no variation ?in the lateral direction?. A typical finite-element mesh with quadrilateral soil elements and beam– column structural elements is shown in Fig. 3. Preliminary ma

28、te- rial properties for use as input in the finite-element model were obtained from the design specifications. After the experimental test, the actual material properties, determined from the test unit through concrete c

29、ore, steel reinforcement, and soil testing, were used to update the finite-element model. To simulate realistic field conditions, the analysis was accom- plished in a step-by-step procedure. Accordingly, the system was m

30、odeled during the different incremental phases of installation, backfilling, and vehicular live load applications. A series of soil lifts was added alongside the bridge unit, representing the actual installation process.

31、 This incremental construction approach is continued, adding one layer of elements at a time, until the pro- cess of backfilling around and over the structure is complete ?see Fig. 3?. Subsequently, the vehicular traffic

32、 loads are simulated in the analysis by applying wheel loads on the surface of the top layer of soil.Soil PropertiesThe validation of assumptions regarding the soil material proper- ties and soil–structure interaction me

33、chanisms is essential for the acceptance of the analysis and design of a buried structure. Thus, the critical input parameters in a finite-element model are typi- cally associated with the soil properties. The subsurface

34、 soil ex- ploration of the bridge site revealed bedrock at the foundation level, consisting of the Richmond formation in the Miami Valley area with alternating layers of shale and hard limestone. The rel- evant parameter

35、s employed in the finite-element analysis for the foundation bed consisted of the bedrock’s engineering properties.Soil ModelThe analytical response of the test bridge was compared with respect to five soil models in the

36、 finite-element program, using the field-determined soil properties as input. The soil models consid- ered were: isotropic linear elastic, orthotropic linear elastic, over- burden dependent, Duncan, and Selig models. Sin

37、ce soil is gen- erally neither a linear–elastic nor a constant-stiffness material, the two linear elastic soil models expectedly gave a poor correlation. The overburden dependent soil model varies soil stiffness with ove

38、rburden pressure, but only for one-dimensional compression of the soil. In zones where two-dimensional compression is prevalent, such as near culverts or walls, this model is not accu- rate ?Katona et al. 1976?. Thus, it

39、 gave a poor correlation as well. As anticipated, better correlation was obtained with the last two models. The Duncan ?Duncan et al. 1980? and Selig ?Selig 1988? models are nonlinear, and have sets of parameters to repr

40、esent a range of soil types. After a number of finite-element runs, it be- came evident that the Duncan and Selig models represented the project at hand more closely. The Duncan model uses variable Young’s and bulk modul

41、i. Both of these moduli increase with confining stress, and Young’s modulus decreases with increasing shear stress. The Selig model is an extension of the Duncan model, using an alternate formulation for the bulk modulus

42、 that is hyperbolic ?Lin 1987; Yang 1987?. The hyperbolic soil model is based on the triaxial compression test in which the soil sample is initially confined by a hydrostatic pressure, ?3. Subsequently, the axial stress,

43、 known as the deviator stress ??=?1??3 is increased until shear failure occurs. Designating the ratio of the actual fail- ure deviator stress and the ultimate deviator stress at large strain as the failure ratio, Rf, Sel

44、ig developed the following transformed equation ?stress–strain relationship? of a hyperbolic form:?? = ?1 ? ?3 = ??? 1Ei + ?????u? ?3?In the above equation, the initial tangent modulus, Ei, is depen- dent on the confinin

45、g pressure and can be represented byEi = KPa? ?3 Pa?n ?4?The atmospheric pressure, Pa, is utilized only to nondimension- alize the modulus number, K, and modulus exponent, n, in the above equation. Furthermore, the bulk

46、modulus, B, defined as the ratio of change in mean stress, ??m, and change in volumetric strain, ??vol, is determined from triaxial dataFig. 3. Finite-element model246 / JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES &