bentley structural the sutong bridge analysis hi res

7/28/2019 Bentley Structural the Sutong Bridge Analysis Hi Res

1/8www.bentley.c

Transit Power Generatio

Communications

T St BiA Structural Analysis

SummaryThe Sutong cable-stayed bridge is the Primary Fairway Bridge o the Suzhou-Nantong Yangtze River Bridge Project in China. At a costo approximately US $920 million dollars, it is an important project with a goal o reducing the economic gap between Suzhou andNantong city and promoting balanced development in the area. Completed in the summer o 2007, the Sutong Bridge is the longestcable-stayed bridge in the world. The total length o the cable-stayed portion o the project is 2,088 meters with a 1,088-meter mainspan and a pylon height o about 300 meters. This paper briey describes the project and discusses some o the structural design andstatic analyses that were undertaken in the design o Sutong Bridge.

InTroducTIon

The Suzhou-Nantong Yangtze River Bridge Project is located in Chinas Jiangsu Province a ast-growing industrial region with apopulation o 74 million people. In recent times, Southern Jiangsu Province has developed rapidly, but the presence o the Yang-tze River has restricted access to the northern portion o the province, limiting its development. The Sutong Bridge will provide animportant link between the cities o Suzhou and Nantong and assist in driving toward the ultimate goals o eliminating poverty andaccelerating mutual prosperity.

The total length o the bridge portion o the project is 8.2 kilometers. The bridge comprises a Primary Fairway Bridge (the Sutong

Bridge), a Special Fairway Bridge, and both approach spans. The Primary Fairway Bridge is a cable-stayed bridge, and the SpecialFairway Bridge is a pre-stressed concrete continuous rigid-rame bridge with a span arrangement o 140+268+140 (548) meters. Theapproach spans are pre-stressed concrete continuous girder bridges 75 meters, 50 meters, and 30 (155) meters in span length.At the bridge site, there are two navigation channels, the Primary Fairway and the Special Fairway or the exclusive use o the porto Nantong.

Chinese ofcials take pride in the act that this enormously challenging project has been an all-China eort without internationalassistance in abrication or construction. Main contractors or management and construction were the Jiangsu Sutong BridgeConstruction Commanding Department (project management) and the China Harbor Engineering Company Group (construction).


2/8www.bentley.c


Communications

Chinas Highway Planning and Design Institute (HPDI) Consultants, Inc. designed the bridge in cooperation with Jiangsu ProvincialCommunication Planning & Design Institute and the Architectural Design & Research Institute o Tongji University. Severalinternational companies served as consultants or special tasks in the design and planning processes. These frms included COWIConsultants and CHODAI Co. Ltd., which independently reviewed the design documents.

Permission granted or oreign companies to contribute to special tasks enabled HPDI to deploy RM Bridge sotware. Dorian

Janjic, vice president o Bentley Sotwares bridge engineering group, supported the HPDI design team in its use o RM duringthe design process.

HPDI chose Bentley and its RM sotware based on the sotwares proven versatility and the Bentley bridge engineering groups experi-enced and solution-oriented development and consulting team. The versatility o Bentleys RM sotware had been demonstratedin its extensive use on Hong Kongs Stonecutters Bridge, the frst cable-stayed bridge to surpass the theretoore accepted limit o1,000 meters or a main cable-stayed span.

This gave HPDI confdence that all obstacles, even those unanticipated, would be overcome through the joint endeavor. This trustwas justifed; in spite o signifcant project challenges and the owners stringent demands, the team met the July 2004 deadline tocomplete the detailed design.

ProjecT challengeSVarious environmental actors and operational demands posed extraordinary design, analysis, and construction challenges:

nviti qits Large container ships and large-scale eets pass under the bridge regularly. Clearance or theshipping lane required a width greater than 891 meters and a height greater than 62 meters. Moreover, the main bridge had tobe designed to resist the impact o a 50,000-ton ship.

P it Each year the region averages 30 days o heavy og, more than 120 days o heavy rain, and high wind speedscaused by typhoons and tornados. As a result, construction teams had to ollow aggressive schedules to complete work inseasonal windows.

cpx Because the river is tidal, it has varying ow speeds, directions, and depths. Waves sometimes reach3 meters in height and currents can be quite strong. Tides can vary by up to 4 meters. The design allowed or an average owpassing through the river cross section o 4.1 meters per second.

dp b The bedrock is at a depth o 270 meters and is covered by sediment, sand, and silt. This meant that a specialsolution was needed or the oundations one that didnt involve drilling into bedrock.

These challenges required sophisticated analyses that studied large displacements caused by many dierent potential conditions.Particularly important were studies o the dynamic behavior resulting rom wind impacts, seismic events, and ship collisions withthe pylons. Analyses were perormed or the ull construction sequence, with special emphasis being placed on optimizing the cabletensioning that is essential at every stage o cable-stayed bridge construction. The analyses included:

optiizti cb Fs For cable-stayed bridges, cable tension is fne-tuned to achieve ideal internal orce distribu-tion in the completed structure. Generally, in such projects, the ideal fnal state is predetermined by basic conditions such asminimal bending moments in the deck and pylons under permanent loads. These criteria govern the strategy o cable tensionadjustment. AddCon, a special RM unction, automatically calculated the optimal distribution o cable orces and the required

cable stressing sequence.

cstti St asis Forward analysis with the AddCon module was used or all erection stages to achieve theoptimal fnal dead load situation required by the designer. The analysis model included a wide range o conditions or variousconstruction stages. The project team also investigated equivalent static wind actions rom dierent directions at constructionphases deemed to be the most problematical.

l dispts The project team paid particular attention to geometric nonlinearities throughout the process. Engi-neers conducted a special study on nonlinear eects the results o which provided notable characteristics o the inuence o


3/8www.bentley.c


Communications

geometrical nonlinearities. Each stay cable was divided into specially developed catenary elements in order to consider cablesagging more accurately, rather than using Youngs modulus or an approximation. Comparisons show that or cable-stayedbridges longer than 1,000 meters, the catenary approach is essential to achieve the required accuracy.

In the construction phase, large deviations rom the design shape have to be adopted as pre-camber values to get therequired design shape under permanent load at the end o the construction sequence without allowing prohibited

internal constraint orces.

Wi Ipt The project team perormed several investigations to gauge the impact o heavy winds:

1. The team developed a suitable girder cross section that satisfes operational demands and bearing capacity requirementsas well as the wind-loading requirements. Wind tunnel tests perormed at Tongji University led to a streamlined, closedsteel box girder with wind airings.

2. The team studied cable vibration caused by wind and rain or periodic excitation. Using RM, the team investigated dierentmethods or minimizing stayed-cable vibrations.

3. The team perormed ull dynamic wind-bueting analyses o the bridge structure with and without trafc. These analyseswere based on aerodynamic coefcients and other data derived rom the wind tunnel tests. Analyses included nonlineardamper elements required or cable stabilization and the girder/pylon connections.

Dynamic Behavior Large displacements, oten caused by temperature changes, can occur in these types o structures duringconstruction as well as in service. During analysis, these movements must not be constrained in order to avoid overstraining. Forthis project, nonlinear dampers were applied or this purpose and or the dynamic loadings. However, these dampers did not con-fne displacements caused by natural conditions. Defning appropriate characteristic design parameters o the damper elements including gap value, elastic stiness, and dynamic characteristics was essential. RM was used to perorm the requiredparametrical studies to design the layout o these devices. The dynamic parameters were based on time-history analysis resultsor some typical seismic inputs.

deScrIPTIon oF The BrIdge

Sp ats

Ater considering various geotechnical conditions at the bridge site, including technical easibility and constructability, the projectteam opted or a double-plane, twin-pylon, cable-stayed bridge design or the Primary Fairway Bridge with a continuous span arrange-ment o 2,088 meters (100+100+300+1,088+300+100+100), as shown in Figure 1. Two auxiliary piers and one transitional pier wereerected in each side span. The main span o the bridge is 1,088 meters, which is the worlds longest main cable-stayed bridge span atthe present time.

Fig. 1: Span Arrangement (unit: m)


4/8www.bentley.c


Communications

gi

The bridge girder is a streamlined, closed, at, steel-box girder. The total width, including wind airing, is 41.0 meters to accom-modate eight dual-trafc lanes. The cross section height is 4.0 meters. The steel-box girder is generally stiened in the longitudinaldirection with closed steel troughs. Transverse plate diaphragms are provided with a typical distance o 4.0 meters and with smallerdistances down to 2.27 meters locally around the two pylons. The characteristic yield strengths o the structural steel are 345 MPa

and 370 MPa. (MPa is a metric unit o measure or steel strength.)

Figure 2 illustrates the standard cross section o the girder. The thickness o the skirts and stieners vary along the longitudinaldirection o the bridge.

Fig. 2: Cross Section of th Girder (unit: m)

Ps

The two 300-meter-tall, inverted Y-shaped pylons are constructed o concrete grade C50 to comply with Chinese standard JTJ01-89. The pylons hold 36-tonne steel boxes astened to the concrete by shear studs at the top o the pylon to anchor the stay cables.Tie beams between the pylon legs are ully post-tensioned to gain an outward thrust rom the pylon legs under service and seismicloads. According to project specifcations and review comments by COWI Consultants, the cracking width o the concrete pylon wall is

controlled within 0.2 millimeters.

St cbs

The stay cables are arranged in double inclined cable planes with a standard spacing o 16 meters in the central span and 12 metersnear the ends o the back spans along the girder. To reduce wind-load eect, the cable-stay systems are made o the parallel wirestrand consisting o 7 millimeter wires, each with a cross sectional area o 38.48 millimeters2. The nominal tensile strength o thecables is 1,770 MPa. Cable sizes range rom PES7-139 or the main span stays near the pylons to PES7-313 or the longest back stay.The longest cable is about 577 meters and weighs 59 tonnes.

During the design process, the project team studied the cable vibration issue caused by wind in combination with rain or parametricexcitation. The project team investigated dierent ways o minimizing stay cable vibrations, including two kinds o cable suracetreatments to prevent rainwater ows rom orming on the cables and internal or additional external damping devices. The fnal mea-sures would be chosen ater detailed testing.

Ftis

Bored riction piles support the piers and pylons rom pier 1 to pier 8 with diameters rom 2.8 meters near the pile-head to 2.5 metersrom the top along the piles. Piers 1, 2, 7, and 8 each have 19 piles, while piers 3 and 6 each have 36 piles, each driven separately.The pylons or pier 4 and pier 5 are supported by 131 piles, varying in length rom 108 to 116 meters.


5/8www.bentley.c


Communications

cti Btw gi Ps

The project team used the same type o nonlinear dampers usedon the Great Belt East Bridge in Denmark to select the permanentconnection between the girder and the pylons. These dampers do notconfne the displacement o the steel girder induced by temperature,

moderate wind, and vehicle trafc, but instead transer the loadsinduced by gusts, earthquakes, and other orces rom specifc loadcombinations rom the girder to an alternative pylon.

The dynamic characteristics o one damper is described by theormula F=CV

V is the relative displacement velocity between pylon and girder is a constant exponent equal to 0.4 C is a constant equal to 3,750 kN/(m/s)0.4

Four dampers were placed at each pylon with a maximum relativedisplacement between the girder and the pylons o less than 750

millimeters to meet the design requirements. Each o the dampersat one pylon has a linear stiness o 100 MN/m to guard against arelative movement beyond 750 millimeters. Figure 3 shows the staticorce-displacement relationship or each damper unit.

gloBal STaTIc analySIS

gb ati m

The HPDI team used RM Bridge1 or the global analysis o the Sutong cable-stayed bridge during detailed design. The designers alsoused QJX and BAP programs or design checking. Figure 4 illustrates the fnite element model o the bridge. The structural modelingo stays was perormed in accordance with the planned construction schemes. Each o the stay-cables was divided into eight sub-

elements to consider cable-sag eects rather than approximating this eect by using an eective module o elasticity. Otherinteracting nonlinear eects such as the P-delta eect, large displacements, and shear displacements were considered in thecalculation. Creep and shrinkage eects were calculated according to the CEB/FIP 90 code. The exibility o the pylon oundationswas modeled with spring elements. The connections between the girder and both pylons were treated as nonlinear static springelements with a gap value o 750 millimeters and a linear stiness o 100 MN/m.

Fig. 4: Finite element model of the bridge

dfiti t Bis Fi Stt ati rsts

Cable-stayed bridges distribute internally the orces in the completed structure using very specifc adjustments in cable tensioning.Cable orce distributions are designed to minimize or even eliminate bending moments in the deck and pylons under permanent loadswhile at the same time avoiding dramatic variations between any two adjacent cables.

Fig. 3: Static force-displacement relationship for each damper


6/8www.bentley.c


Communications

The contribution ratio o trafc loading or the Sutong Bridge was heavy or deck stress and counterweight arrangements in the backspans. The defnition o the fnal state took into account situations with and without trafc. Figure 5 shows the bending momentenvelopes in the deck under dead load and load combination. At 28.0 MN/m, the maximum moment o the pylons (shown in the graphin Figure 5) was very minor. The results illustrate the suitability o the achieved fnal state o the bridge.

Fig. 5: Bending moment envelope in deck

St asis

Using RM ADDCONs orward analysis method2 or the erection stages, the team was able to match the fnal-state conditions de-scribed above as derived rom the original construction schedules o the designer. The analysis model included, at various stages,all temporary supports, tie-downs, and derrick movements or construction, temporary loading, and permanent loading. The team alsoinvestigated equivalent static wind actions rom dierent directions at the most critical construction stages the maximum doublecantilever, the maximum single cantilever, and the bridge completion stage.

RM automatically computed the pre-camber o all construction. With the exception o the design elevation o the deck, the third-ordereect o pre-camber shapes was taken into consideration. Results rom construction stage analysis showed a minimal amount ostiness beore closure. For instance, initial tensioning o the longest stay cable in mid-span yielded a vertical deection o 1.3meters at the end o the cantilever. Even ater closure, the superimposed dead load (including paving, barriers, etc.) still yielded avertical deection o 1.8 meters in the center o the mid-span. The results clearly demonstrate the benefts o RMs geometric nonlin-earity analyses, especially concerning the deck erection geometry.

Stt Sst Pti ass

As mentioned, the dampers do not confne the displacement o the steel girder induced by temperature, moderate winds, and vehicletrafc, but transer the loads o the girder induced by gusts, earthquakes, and other orces to the tower. Thereore, defning appropri-ate characteristic design parameters o the dampers including gap value, elastic stiness, and dynamic characteristics is criticalto achieve the desired results. Relevant parametrical analyses were carried out or some dominant load cases, including staticloadings and dynamic inputs.

The dynamic parameters are based on the results o time history analyses or some typical seismic inputs.

For static actions, a proper gap value was the governing parameter. Taking into account all o the responses o the above loads andconsidering the current product specifcations o large expansion joints, a gap o 750 mm was selected to ft the design requirements.Consequently, some parametrical analyses were carried out to defne reasonable spring stiness according to the interaction curvebetween the bending moment response at pylon bottom (the longitudinal displacement at the end o the girder) and longitudinalwind inputs.


7/8www.bentley.c


Communications

Figures 6-8 show results o the parametrical studies with the spring stiness in the range o 1 to 1000MN/m. These graphs illustratethat a spring stiness o 100MN/m is practicable. An assumption was introduced here that only the dampers at one pylon o pier4 or pier 5 would be engaged beore the other pylon. This assumption reects construction inaccuracy and bridge deck elongationdue to temperature.

For the Sutong Bridge, the connection between girder and pylons is essential or the saety o the pylons under extreme wind andseismic loads. Thereore, based on the perormed detailed parametrical studies and some urther considerations or installationtolerances and saety margins, a maximum orce or one damper o approximately 10 MN (under ULS state) was assumed as one othe design prerequisites. Meanwhile, the comparison between the results o HPDI and COWI Consultants confrmed that the materialnonlinearity o the pylons plays a signifcant role in the resultant reaction orces under ULS state.

gti niit ets

Designers gave much attention to geometric nonlinearities all the way rom the preliminary design phases to the detaileddesign. The designers carried out a special study on nonlinear eects3. Two notable remarks on geometric nonlinearitiesare abstracted as ollows:

Compared with linear analysis, eects o geometrical nonlinearity may result in a net oset o 10 to 20 percent o the maximum/minimum stress o the girder and the pylons together with a shit in the critical location o these stresses.

Generally, the fnite element model o the stay cables employs a straight truss element with the eective modulus o elasticity,or using RM to divide each stay cable into many sub-elements, or by the new catenary cable element. Dierent means to dealwith cable-sagging eects result in various options or the abrication or construction processes. The means o the equivalenttruss element may induce a maximum oset o 0.538 meters away rom the desired location at the end o stage analyses4. Oneo the possible reasons is inaccurate chord-orce vectors in long-stay cables. For cable-stayed bridges longer than 1,000 meters,this simplifcation should be restricted, especially or erection processes. Certainly, the use o catenary cable elements is betterthan dividing sub-elements, but within the tolerance range.

concludIng remarkS

Certainly, the Sutong Bridge is an amazing eat o design and engineering. The design and construction o the bridge itsel has pro-vided a very good opportunity to promote cooperation and interaction among many amous bridge designers rom China and abroad all the more appropriate in this ast-growing area o the world. Most important, the bridge brings a new level o convenience to thepeople along the Yangtze River and should accelerate economic development and mutual prosperity within the cities it bridges.

Fig. 6: Maximum moment at pylon bottom Fig. 7: Displacement at pyon top Fig. 8: Total reaction force of four dampers


8/8www.bentley.c


Communications

reFerenceS

1. TDV, 2008, RM User Guide, rev 9.51.10.

2. Janjic D., Pircher M., Pircher H. Optimization o Cable Tensioning in Cable-Stayed Bridges, Journal o Bridge Engineering, ASCE,v8, n3, pp 131-137, 2003

3. MIAO Jiawu, Comparison Study on Geometrical Nonlinearity Eects or the Sutong Cable-Stayed Bridge. 2003 Conerenceproceedings o Bridge and Structure subdivision o Chinese Highway Association. Sep, 2003.

4. LIANG P., Geometrical Nonlinearity and Random Simulation o Super Long Span Cable-Stayed Bridges. A dissertation or thedegree o Doctor o Philosophy o Tongji University. Aug, 2004.

2009 Bentley Systems, Incorporated. Bentley, and the B Bentley logo are either registered or unregistered trademarks or service marks of Bentley Systems, Incorporated or one of its direct or indirect wholly ownedsubisdiaries. Other brands and product names are trademarks of their respective owners. BAA016750-1/0001 08/09

bentley structural the sutong bridge analysis hi res

Documents