Tài liệu Congress on technical advancement 2017 cold regions engineering

Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. Congress on Technical Advancement 2017 Cold Regions Engineering Proceedings of the Congress on Technical Advancement 2017 Duluth, Minnesota September 10–13, 2017 Edited by Jon E. Zufelt, Ph.D., P.E., D.WRE Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. CONGRESS ON TECHNICAL ADVANCEMENT 2017 COLD REGIONS ENGINEERING PROCEEDINGS OF THE 17TH INTERNATIONAL CONFERENCE ON COLD REGIONS ENGINEERING PRESENTED AT THE FIRST CONGRESS ON TECHNICAL ADVANCEMENT September 10–13, 2017 Duluth, Minnesota SPONSORED BY Committee on Technical Advancement Aerospace Engineering Division Cold Regions Engineering Division Committee on Adaptation to a Changing Climate Energy Division Forensic Engineering Division Infrastructure Resilience Division Construction Institute Duluth Section of ASCE Utility Engineering and Surveying Institute of the American Society of Civil Engineers EDITED BY Jon E. Zufelt, Ph.D., P.E., D.WRE Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. Published by American Society of Civil Engineers 1801 Alexander Bell Drive Reston, Virginia, 20191-4382 www.asce.org/publications | ascelibrary.org Any statements expressed in these materials are those of the individual authors and do not necessarily represent the views of ASCE, which takes no responsibility for any statement made herein. No reference made in this publication to any specific method, product, process, or service constitutes or implies an endorsement, recommendation, or warranty thereof by ASCE. The materials are for general information only and do not represent a standard of ASCE, nor are they intended as a reference in purchase specifications, contracts, regulations, statutes, or any other legal document. ASCE makes no representation or warranty of any kind, whether express or implied, concerning the accuracy, completeness, suitability, or utility of any information, apparatus, product, or process discussed in this publication, and assumes no liability therefor. The information contained in these materials should not be used without first securing competent advice with respect to its suitability for any general or specific application. Anyone utilizing such information assumes all liability arising from such use, including but not limited to infringement of any patent or patents. ASCE and American Society of Civil Engineers—Registered in U.S. Patent and Trademark Office. Photocopies and permissions. Permission to photocopy or reproduce material from ASCE publications can be requested by sending an e-mail to [email protected] or by locating a title in ASCE's Civil Engineering Database (http://cedb.asce.org) or ASCE Library (http://ascelibrary.org) and using the “Permissions” link. Errata: Errata, if any, can be found at https://doi.org/10.1061/9780784481011 Copyright © 2017 by the American Society of Civil Engineers. All Rights Reserved. ISBN 978-0-7844-8101-1 (PDF) Manufactured in the United States of America. Congress on Technical Advancement 2017 iii Preface Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. The Congress on Technical Advancement was established to bring together several of the Divisions under the ASCE Board-level Committee on Technical Advancement (CTA) at a single venue. While some of the CTA Divisions hold regular small conferences, others do not have an established forum to present technical information to their constituents or the engineering community. One of the goals of the Congress is to provide greater opportunities for interaction and synergy among the activities of the Divisions and ASCE’s Institutes. This 1st Congress on Technical Advancement was held at the Duluth Entertainment and Convention Center in Duluth, Minnesota on September 10-13, 2017. This 1st Congress included the participation of and presentations by the Aerospace Engineering Division, Cold Regions Engineering Division, Committee on Adaptation to a Changing Climate, Energy Division, Forensic Engineering Division, Infrastructure Resilience Division, the Construction Institute (CI), and the Utilities Engineering and Surveying Institute (UESI), representing the combination of existing conference series as well as opportunities for new periodic technical symposia. The Congress was hosted by the Duluth Section of ASCE as they celebrated their 100th Anniversary with a special session and evening social event. The 2017 Congress on Technical Advancement included 3 days of presentations with daily plenary sessions followed by 6 parallel tracks of technical sessions providing a venue for over 160 presentations. The conference also included an Awards Luncheon highlighted by the presentation of the Harold R. Peyton Award for Cold Regions Engineering, the CANAM Civil Engineering Amity Award, the Charles Martin Duke Lifeline Earthquake Engineering Award and the Alfredo Ang Award on Risk Analysis and Management of Civil Infrastructure. Other recognitions during the Congress include the Eb Rice Lecture Award, the Best Journal of Cold Regions Engineering Paper Award, and the Best Cold Regions Conference Paper Award. An Opening Congress Reception, Duluth Section 100th Anniversary Session and Social Event, and Technical Tours provided additional opportunities for attendees to share ideas. This collection of 60 papers brings together the current state of knowledge on a variety of topic areas presented at the 2017 Congress on Technical Advancement and is separated into three EBooks. The first represents selected papers from the Proceedings of the 17th International Conference on Cold Regions Engineering. The second includes the papers on Infrastructure Resilience, Aerospace and Energy. The third EBook presents papers addressing Construction and Forensic Engineering. I would like to thank all of the volunteers and ASCE Staff who have made this 1st Congress on Technical Advancement and Proceedings possible. It could not have been done without all of the authors, reviewers, attendees, and Congress Committee members. Jon E. Zufelt, Ph.D., PE, D.WRE, F.ASCE Congress Chair and Proceedings Editor © ASCE Congress on Technical Advancement 2017 Acknowledgments Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. Congress Organizing Committee Jon Zufelt, Ph.D, P.E., CFM, D.WRE, F.ASCE James Anspach, P.G. (ret.), F.ASCE Ron Anthony, Aff.M.ASCE Hiba Baroud, Ph.D., Aff.M.ASCE Ana Boras, Ph.D, P.E., M.ASCE Martin Derby, A.M.ASCE Mike Drerup, P.E., M.ASCE Jim Harris, P.E., Ph.D, F.SEI, F.ASCE, NAE John Hinzmann, P.E., M.ASCE Jen Irish, Ph.D, P.E., D.CE, M.ASCE John Koppelman, A.M.ASCE Tom Krzewinski, P.E., D.GE, F.ASCE Bob Lisi, P.E., M.ASCE Juanyu "Jenny" Liu, Ph.D., P.E., M.ASCE Pat McCormick, P.E., S.E., F.ASCE, F.SEI Nick Patterson, P.E., M.ASCE David Prusak, P.E., M.ASCE Ziad Salameh, P.E., M.ASCE J. "Greg" Soules, P.E., S.E., P.Eng, SECB, F.SEI, F.ASCE Amy Thorson, P.E., F.ASCE Nasim Uddin, P.E., F.ASCE Joel Ulring, P.E., M.ASCE ASCE Staff Susan Davis, A.M.ASCE Jon Esslinger, PE, F.ASCE, CAE Mark Gable Katerina Lachinova Shingai Marandure Amanda Rushing, Aff.M.ASCE Jay Snyder, Aff.M.ASCE Catherine Tehan, Aff.M.ASCE Drew Caracciolo © ASCE iv Congress on Technical Advancement 2017 v Proceedings Reviewers Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. Il-Sang Ahn Lorenzo Allievi Ron Anthony Navid Attary Bilal Ayyub Eugene Balter Heather Brooks Henry Burton ZhiQiang Chen Adrian Chowdhury Edwin Clarke Billy Connor Craig Davis An Deng Alicia Diaz de Leon Curt Edwards Jon Esslinger Caroline Field Madeleine Flint Chris Ford Warren French Subhrendu Gangopadhyay Rob Goldberg Scott Hamel John Henning Jiong Hu Baoshan Huang © ASCE Josh Huang Joshua Kardon Mehrshad Ketabdar John Koppelman Thomas Krzewinski David Lanning Spencer Lee Jenny Liu Hongyan Ma Rajib Mallik Tony Massari Roberts McMullin Ralph Moon Anthony Mullin Mark Musial LeAnne Napolillo Kevin Orban Sivan Parameswaran Tim Parker Robert Perkins Brian Phillips Chris Poland Allison Pyrch Craig Ruyle Bill Ryan Stephan Saboundjian Ziad Salameh Andrea Schokker Yasaman Shahtaheri Jim Sheahan Xiang Shu John Smith Ryan Solnosky Greg Soules Bucky Tart Scott Tezak Ganesh Thiagarajan Eric Thornley John Thornley Jeff Travis Nasim Uddin Joel Ulring Shane Underwood Cindy Voigt Dan Walker Haizhong Wang Chenglin Wu Gang Xu Zhaohui Yang Kent Yu John Zarling Chris Zawislak Weiguang Zhang Jon Zufelt Congress on Technical Advancement 2017 vi Contents Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. Effect of the Presence of Pre-Service Construction Cracks in Concrete Decks on the Thermal Profile of Composite Steel-Concrete Bridges in Cold Regions ............................................................................................. 1 Omar Y. El Masri and Caesar Abi Shdid Opportunities for Increased Utilization of Geothermal Resources in the United States.............................................................................................................. 13 Gretchen E. Schimelpfenig Effect of Diamond Grinding on Pervious Concrete............................................... 25 B. I. Izevbekhai Effects of Aggregate Thermal Contraction Properties on Asphalt Mixture’s Thermal Properties and Low Temperature Performance .................................... 37 Moses Akentuna and Sang-Soo Kim Bagged Reinforced Concrete Shaft in Saline Soils in Cold Regions .................... 50 Yu Zhang, Jianhong Fang, Zhaohui (Joey) Yang, and Jiankun Liu The Use of Electrical Resistivity Methods for Ground Ice Characterization for Engineering ......................................................................................................... 59 Kevin Bjella, Misha Kanevskiy, and Kenneth Hinkel Internal Observation of Soil in Frost Heave Process Using the X-Ray CT Scan...................................................................................................................... 71 Baiyang Song, Dai Nakamura, Takayuki Kawaguchi, Shunzo Kawajiri, Satoshi Yamashita, and Dahu Rui Haughton-Mars Project at 20: Challenges and Designs for Future Exploration ................................................................................................... 79 Sarah J. Seitz and Brian J. Glass Evaluation of a Simulated Roadway Weather Information System’s Pavement Temperature Data in Illinois ................................................................. 91 Wouter Brink, Carmine Dwyer, Gregg Larson, and William Vavrik Methods and Quality of Embankments Constructed in Winter ........................ 102 Atsuko Sato, Toshihiro Hayashi, Teruyuki Susuki, and Shinichiro Kawabata Numerical Modeling of Electrically Conductive Pavement Systems ................. 111 S. M. Sajed Sadati, Kristen Cetin, and Halil Ceylan © ASCE Congress on Technical Advancement 2017 Frost Resistance of High Early-Age Strength Concretes for Rapid Repair...... 121 Matthew Maler, Meysam Najimi, and Nader Ghafoori Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. Frost Resistance of Self-Consolidating Concrete Containing Natural Pozzolan ................................................................................................................... 132 Meysam Najimi, Mohammad Reza Sharbaf, and Nader Ghafoori Estimating Liquefaction Potential of Thawing Permafrost Soils, an Evaluation of Frozen and Recently Thawed Soils ............................................... 141 John D. Thornley, Nick Moran, Andrew Daggett, and Thomas G. Krzewinski Thermosyphon Design for a Changing Arctic ..................................................... 151 Edward Yarmak and Jason T. Zottola Stress Analysis of the Phoenix Compacted Snow Runway to Support Wheeled Aircraft .................................................................................................... 161 Ariana Sopher and Sally Shoop Thermal Evaluation of Common Locations of Heat Loss in Sandwich Wall Panels .............................................................................................................. 173 Taylor Sorensen, Sattar Dorafshan, and Marc Maguire Wear of Steel and Heavy Duty Coating Caused by Friction of Sea Ice ............. 185 S. Kioka, N. Maruta, and T. Takeuchi On the Thickness of Sacrificial Steel Wall for Protection of Sea Ice Action ....................................................................................................................... 195 T. Takeuchi, S. Kioka, and H. Miyazaki Comparing Ground Snow Load Prediction Methods in Utah............................ 207 Brennan Bean, Marc Maguire, and Yan Sun © ASCE vii Congress on Technical Advancement 2017 Effect of the Presence of Pre-Service Construction Cracks in Concrete Decks on the Thermal Profile of Composite Steel-Concrete Bridges in Cold Regions Omar Y. El Masri1 and Caesar Abi Shdid2 Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. 1 Ph.D. Candidate, Dept. of Civil and Environmental Engineering, Syracuse Univ., Syracuse, NY 13210. E-mail: [email protected] 2 Associate Professor and Chair, Dept. of Civil and Environmental Engineering, Lebanese American Univ., 211 E 46th St., New York, NY 10017. E-mail: [email protected] Abstract Thermally induced stresses in composite steel-concrete bridges are higher than those experienced by their concrete and steel cousins, leading to significant damage in the concrete deck and corrosion of the steel reinforcement. Bridge design engineers use thermal profiles prescribed by codes such as AASHTO to predict future service stresses. A 3D finite element model is presented that investigates the temperature distribution in a case study bridge with pre-existing construction deck cracks. The non-linear transient simulation is performed using actual environmental loads for a geographic region with severe climate (North Dakota), and the resulting profile is compared to that of AASHTO. The results show the thermal gradient proposed by AASHTO to be overly conservative in cold regions. Existing models seem to ignore the nonlinearity of the thermal gradient, which can be critical for thermal stress calculations. The pre-service deck cracks appear to have a considerable effect on both the vertical and the longitudinal temperature distributions, and it is recommended that they be given careful consideration by design codes. 1. Introduction and Justification Bridges are subjected to continuously changing diurnal environmental conditions that lead to continuous heat gain and loss with their surroundings. The thermal gradient that develops within a bridge cross section is affected by four basic heat transfer phenomena: a- convection at the surfaces, b- irradiation, c- solar radiation, and d- conduction within the bridge. While the solar radiation intensity has the highest effect on changing the bridge temperature, the thermal gradient is largely affected by the thermal diffusivity of the constituent materials. It is the difference in this thermal diffusivity of concrete and steel that makes the thermal gradient in composite steelconcrete bridges rather high. The non-uniform temperature distribution within a bridge cross section when combined with different coefficients of thermal expansion and shear connectors that prevent slip between the concrete deck and steel girders, will lead to considerable thermal stresses. These stresses are known to be relatively high when compared to service load stresses, leading to considerable damage in the concrete deck. The major damage attributed to thermal stresses is developing deck cracks. However, and despite its importance, limited studies have been dedicated to investigating the temperature distribution in composite bridges [1,2,3]. Design codes, such as the American Association for State Highway and Transportation Officials (AASHTO) LRFD Bridge Design Specifications, assert the importance of accounting for thermal stresses in bridge design by providing designers with proposed thermal gradients that describe © ASCE 1 Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. Congress on Technical Advancement 2017 the vertical temperature distribution in bridges located in various geographic regions [4]. However, previous studies, on which the proposed AASHTO gradient is based, mostly consider two- or one-dimensional models with, in some cases, even steady state analysis. They have thus failed to consider the effect of construction cracks, which are pre-existing in the bridge deck, on the temperature distribution within the deck in both the transverse and longitudinal directions. These cracks are found to develop directly after the concrete deck casting and before the opening of the bridge for traffic [5]. The work presented here uses a three-dimensional computational model to examine the accuracy of presently used thermal profile models in cold regions. 2. Related Work Analytical, numerical, and experimental investigations have led to the development of various thermal profiles that have been adopted by different codes around the world. Zuk [1] developed equations to calculate the longitudinal and transverse stresses in composite bridges under different conditions of temperature and shrinkage. These equations were developed for four (4) different and critical cases of temperature distribution; however, a uniform temperature for the steel beam is adopted in all cases due to its high thermal conductivity and its ability to adjust its temperature quickly to that of the surrounding environment [1]. In a later study, Berwanger [6] developed a numerical procedure that uses two-dimensional thermo-elastic finite element analysis (FEA) to predict the transient temperature in the cross sections of composite bridges. Results showed a slower response for the concrete slab with a very rapid increase in thermal moments. The study concluded that a linear temperature profile could be used satisfactorily to represent the temperature in the transverse cross section. The study also stresses that possible existing cracks in the concrete deck were ignored. Thermal gradients used in composite bridges differ from one code to another. Imbsen et al. [2] evaluated the thermal effects on bridge superstructures based on different codes. Many of the findings and recommendations of this study were included in the following revision of the AASHTO code: Thermal Effects in Concrete Bridge Superstructures [7]. Kennedy and Soliman [8] synthesized the various theoretical and experimental studies that had been conducted on composite concrete slab on steel beam bridges, and proposed a simple one dimensional vertical temperature distribution within the section. The distribution they proposed is uniform through the depth of the steel beam and is linear through the concrete deck. A study by Fu et al. [3] concluded that a steady-state thermal condition never exists within a bridge structure, and that the time dependency of the ambient air temperature and solar radiation would dictate a transient analysis. A more recent thermal profile was proposed by Chen [9] based on numerical analysis using two-dimensional finite element (FE) analysis. Emanuel and Taylor [10] conducted a computer-based study on composite bridges to investigate the relationship between uniform, linear, and non-linear components of thermally-induced stresses on the one hand and varying span lengths, number of spans, and support conditions on the other hand. The study concluded that the three components of thermally induced stresses are independent of the span length. Bridge decks with overhangs present a problem for predicting the daily temperature in a cross section due to the shading effect that they will have on the steel girders—an effect that will vary between geographical locations and throughout the time of the day. An analytical parametric study was conducted by Fu et al. [3] on composite bridges to determine the effects of shading. © ASCE 2 Congress on Technical Advancement 2017 Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. The study concluded the shading on the girders from the slab overhang to be the most influential factor on the vertical thermal distribution [3]. Another study confirmed the previous theoretical findings through temperature measurements on experimental bridge scaled models placed on the roof of a building as well as on an existing steel bridge in Hong Kong. [11]. Cracks occur in concrete bridge decks in different forms: transverse cracks, longitudinal cracks, and map cracks. Map cracks (also known as pattern cracks) are a very common form of cracking in all types of concrete bridge decks. Map cracks initiate at the bottom of the concrete deck and propagate their way up through the deck until they immerge at the surface where they appear like a map with squiggly lines. Such cracks are often the product of improper curing where the concrete surface moisture is allowed to evaporate too quickly, and the movement or shrinkage of the concrete deck is restrained. Studies have found transverse cracks (cracks that run perpendicular to the girders) to be the predominant form of cracking in the reinforced concrete bridge decks [5,12]. Transverse cracks have been found to be of full depth of the deck and occur at regular intervals of 0.9 to 3.1 meters apart along the bridge length [13, 14, 15], in both the positive and negative moment regions of the bridge [16]. The widths of the cracks have been reported in the range of 0.1 to 0.5 mm. These cracks have been observed along the entire length of bridges, in both simple and continuous span construction. Ramey et al. [12] noted that transverse cracks occur early during the construction process typically after the casting of the concrete, and before the bridge has been placed in service. 3. Composite Bridge Model The City of Fargo in North Dakota (Latitude: 46˚, 52’, 38”; Longitude: 96˚, 47’, 22”; Elevation: 275 m.), with its extremely cold winters and very warm summers, was selected as an appropriate location for to represent cold regions. This choice is made because such extreme climate highlights the vast thermal differentials that can develop in composite bridges, and consequently produce high thermal stresses. 3.1 Time Domain For any thermal stress calculations for bridges, two cases of temperature conditions must be considered by an engineer in order to estimate the critical stresses for design. The first case is in the summer when the deck is hotter than the steel beams, and the second case is in the winter. The time domain chosen for this transient analysis consists of two 24-hour time spans occurring over two separate days: December 23 and June 4. The selection was based on the lowest and highest radiation intensity days of the year, respectively, for Fargo, ND. The selection was based as such since it has been shown that the solar radiation has the highest influence on the thermal gradients in bridges [11]. This presents an improvement over existing models that have always assumed that the two days for the simulation that represent the critical thermal cases of the year are those with the highest and lowest ambient temperatures, where these temperatures were calculated based on existing sinusoidal empirical expressions [10]. A transient time step of one hour, divided into four (4) equal increments, was used for the thermal simulation. This choice was based on two factors: 1- the fact that weather data used is only available in increments of one (1) hour; and 2- very little temperature, radiation, and wind speed changes occurred over a time span of one hour. © ASCE 3 Congress on Technical Advancement 2017 4 Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. 3.2 Bridge Model Properties Colquitz River Bridge was selected as a case study for its typical nature of many composite steelconcrete bridges, and due to its perfectly symmetrical cross section. The bridge is located near Victoria in British Columbia, Canada. The bridge length is 83 meters divided into five spans of varying lengths, and it has a width of 11.9 m. The concrete deck is 18 cm thick resting on six wide-flange (W33x141) steel girders equally spaced at 198 cm on center with 99 cm overhangs on each side. The material properties that are used in the model are summarized in Table 1. Table 1. Bridge Model Material Properties Bridge Part Concrete Steel Unit Modulus Poisson’s Thermal Coefficient of Weight of Elasticity Ratio Thermal Expansion Conductivity (Kg/m3) (MPa) (W/m K) (cm/cm/oC) 2,400 7,850 26,435 200,000 0.2 0.3 10.8 E-6 11.7 E-6 0.7 42.5 Grade Compressive Strength (MPa) NA ASTM A992 28 NA Given the bridge cross-sectional symmetry with respect to the longitudinal axis, only a portion of the bridge cross section 4 m wide that includes an exterior girder and one adjacent interior girder is needed to develop the thermal profile for the full cross section. Since the thermal stresses are not sought after in this study, the full span and the corresponding support conditions are not necessary to develop the thermal profile, and a 2.75 m longitudinal segment of the bridge was modeled. The post-construction and pre-service transverse cracks in concrete decks were incorporated in the model at a spacing of 0.9 m, a width of 0.51 mm, and extending the full depth of the deck, as reported in the literature. The study assumes that the bridge is neither equipped with a waterproofing system nor with a top asphalt pavement. 3.3 Environmental Loads and Boundary Conditions Bridges are exposed to various environmental variables that lead to heat energy exchange between their surfaces and the surrounding. This exchange is the reason behind the unsteady thermal state within the bridge cross section. The different heat transfer components acting on the boundaries of a bridge are visually depicted in Figure 1 and may be represented in terms of heat flux as = + + , where q is rate of energy transfer, J/s-1 m2 or W/m2; qc is the rate of energy convection; qr is the rate of thermal irradiation, and qs is the rate solar radiation energy. The convection component of the heat flux equation, qc, is calculated using = ℎ ( − ), where hc is the convection heat transfer coefficient, W/m2 ˚C; T is the temperature of the surface, ˚C; and Ta is the ambient temperature, ˚C. The convective heat transfer coefficient is calculated using the empirical formula suggested by Ibrahim [20] as ℎ = 4.67 + 3.83 for top surfaces, ℎ = 2.17 + 3.83 for soffit surfaces and ℎ = 3.67 + 3.83 for steel webs and slab outer surfaces, where u is the wind speed in m/s and hc is in W/m2. The air temperature and the wind speed data for the two chosen days were obtained for the Hector International Airport in Fargo, ND from the National Climatic Data Center [18]. This use of actual ambient temperature and wind speed data for the actual location of the case study bridge represents an improvement over existing models that have exclusively used data synthesized from semi-empirical expressions. The irradiation component in the heat flux equation, qr, is calculated using ( − ), where F is the Stefan-Boltzmann constant equals to 5.671 x 10-8 kg s-3 ˚K-4; T is the temperature of the surface, oC; Ta is the ambient temperature, oC; and is the emissivity of the surface. The © ASCE Congress on Technical Advancement 2017 Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. emissivitty value for concrete c and d rusty steel and iron is rreported by A ASHRAE [119] to be bettween 0.85 and 0.95. Thus, a value of 0.9 0 has been assigned in tthis study foor both materrials. Figurre 1. Heat traansfer process in a bridgge exposed too environmeent The heat radiation co omponent off the heat flu ux equation, q s, is calculaated using , where is the absorptiv vity for solaar radiation; and is th he total houurly solar raadiation on a bridge surrface, Btu/m2. The T absorptiivity value fo or concrete and a rusty steeel and iron iis reported bby ASHRAE E [19] to be bettween 0.65 and a 0.80. A value of 0.8 has been selected to bbe used in tthis study foor the concrete slab and thee steel girderrs. The hourlly total solarr radiation on a bridgee surface conntains three com mponents: beam radiatio on, diffuse radiation, r annd ground-reeflected radiaation. Duffiee and Beckman n [23] pro ovided an expression to computte as = + + , where is the totaal solar radiaation; is tthe beam soolar radiationn on a horizzontal surface; is the difffuse radiatio on on a horizzontal surfacce; I is the tootal radiation on a horizzontal surface ( + ); is i the angle of o incidencee; is the zeenith angle; is the sloppe of the surrface; is the diffused gro ound reflecttion and is equal to 0.22. The expreession to deetermine coss is cos ∅ sin cos +cos cos ∅ cos cos +cos sin ∅ sin n cos coss sin ∅ cos - +cos siin sin sin n , where ∅ is the su urface latitudde; is thee declination; the suurface azimuth angle; is hour angle. The hourly y solar radiattion data at the selectedd location foor the two chossen days, inccluding the beam b and difffused comp onents annd , were oobtained from m the National Renewable Energy Laboratory [21]. c by th he deck overrhang on thee web of the exterior girdder is The hourrly height off the shade created given by = ( , where Lc iis the lengthh of the overrhang slab inn cm; ) is the solar altitud de angle in degrees; d iss the surfacee azimuth anngle in degrees; is thhe sun grees. The hourly h solar altitude anggle and sun aazimuth anggle were obtained azimuth angle in deg from the National Reenewable En nergy Laborratory [21] fo for the choseen bridge loccation and oon the two seleccted days in June and Deecember. Su uch actual meeasured dataa considers tthe actual alttitude of the briidge above th he sea level,, and how th hat affects thee solar anglees. © ASCE 5 Congress on Technical Advancement 2017 Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. 3.4 Finite Element Modeling Studies on the thermal behavior of composite bridges found in the literature have conducted oneand two-dimensional FEA in which the temperature is assumed to remain constant along the length, and sometimes the transverse width, of the bridge [3,23]. The 2D model is widely accepted as an accurate approach for conducting transient heat transfer simulation in composite bridges. However, such 1D and 2D models fall short of being able to model the thermal effect of transverse cracks that have been repeatedly reported to exist prior to the bridge being put into service. A 3D model is used in this study to better reflect the effect of such transverse cracks on the temperature distribution within the bridge cross section. The FE model of the bridge section is constructed using homogenous solid elements. Both concrete deck and steel girders were modeled using 3D 8-node linear hexahedral heat transfer mesh elements with temperature as a single degree of freedom at each node, and second-order (quadratic) interpolation in three dimensions. The refined mesh elements measured 2.25 cm, thus resulting in eight (8) elements along the depth of the concrete deck. The overall model measured 4 m in the transverse direction and 2.75 m in the longitudinal direction, with 0.51 mm wide cracks spaced at 0.9 m in the longitudinal direction and stretching the entire width of the deck. Uncoupled heat transfer analysis was conducted in which the temperature field was calculated without consideration of any stresses or deformations. The nonlinearity in the analysis is the result of the boundary conditions being nonlinear, and the latent heat properties of the elements. The cracks were modeled as an open cavity element with constant size throughout the depth of the deck equal to the width of the crack. The air inside the crack was given the same initial conditions as the air outside the crack. Natural convection and surface irradiation interaction properties are assigned to the two surfaces of each crack. Natural convection at the crack surfaces is generated by density differences in the air occurring due to the temperature gradient across the crack surface. Solar and diffused radiation were not accounted for at the crack surfaces due to its small width and the inability of the radiations to reach such surfaces as shading is provided by the adjacent surface. The contact direction with adjacent elements was modeled to be normal to the vertical surfaces of the crack, and the environmental boundary conditions (arranged in text files) are assigned to the surface elements that recall data from these files at every transient time step. Previous studies have indicated that initial temperature appears to have very little effect on the temperature difference within a bridge deck [3,10]. Given this fact, the bridge temperature was initially set to =0℃, and the environmental initial boundary conditions were set as obtained from the National Climatic Data Center [18]. To ensure a full composite action between the concrete deck and the steel girders with no slip, a “Tie” interaction was used at the contact surface between the two components. This approach ensures a hard pressure over-closure between the two surfaces and hence permits the heat to transfer through conduction. Surface irradiation to the surrounding medium was modeled as a time dependent surface property covering the whole model with uniform emissivity distribution and hourly ambient temperature data. Three different sets of convection heat fluxes (for the top surfaces, soffit surfaces, and side surfaces) were calculated and implemented in the FE model as boundary conditions. © ASCE 6 Congress on Technical Advancement 2017 7 The vertical temperature distribution was obtained at four distinct critical positions within the bridge. These positions are: Position I: midway between the cracks, for the exterior girder. Position II: near the surface of the crack, for the exterior girder. Position III: midway between the cracks, for the interior girder. Position IV: near the surface of the crack, for the interior girder 4.1 Exterior versus Interior Girders An initial comparison of the vertical temperature distributions between the bridge sections at the exterior and interior girders revealed similar results between Positions I and III on the one hand, and Positions II and IV on the other hand. This similarity in the temperature profile was valid for both simulated days. A maximum temperature differential of 0.45 ˚C was reached between the two girders at time 8:00 on June 4, 2010. This small difference is due to the sun beams radiations reaching the exterior steel web soon after sunrise. Yet, the shading effect on the exterior girder for the remainder of the day results in the same temperature distribution for the two girders. 4.2 Absolute Maximum Temperature in the Section 120 Concrete Deck 100 80 15 20 25 30 60 40 Steel Girder 20 0 Temperature (˚C) 45 40 35 30 35 40 25 20 15 Position I 10 5 0 Elevation (in) The absolute maximum temperature in the concrete deck is critical for calculating the bridge thermal stresses resulting from the various components of thermal strains (uniform, linear, and non-linear). On June 4, the temperature in the concrete deck reached a maximum value of 38.6 ˚C at time 15:00 at position I. This temperature was recorded at the top surface of the concrete deck as can be seen from the color contours in Figure 8. At this same time, the temperature at position II reached a maximum value of 33.7 ˚C at the top end of the deck. The reason for this difference of 4.8 ˚C in the maximum temperature between Positions I and II is the presence of cracks, which appear to help the concrete deck surface at mid-depth to quickly adjust its temperature to the transient conditions of the surrounding. This effect is more noticeable in the summer than it is in the winter. Yet, a very high wind speed during the winter could lead to a negative gradient at the crack position (Position II) due to forced air flow through the cracks. Figure 2 provides the vertical temperature distribution at these two locations. It is worth noting that due to solar radiations reflected from the ground, the temperature at the bottom surface of the concrete deck tends to be slightly cooler right over the two steel girders than the regions between the two girders and under the overhang. Elevation (cm) Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. 4. Discussion of Results Figure 2. Vertical temperature distribution at Positions I and II (15:00, June 4) © ASCE Congress on Technical Advancement 2017 8 The absolute maximum temperature in the steel girder will also significantly affect the thermal stresses in the bridge. On June 4, the temperature in the steel girder reached a maximum value of 28.6 ˚C at time 14:00 in the middle of the steel web. The maximum temperature on December 23 peaked at only -4.4 ˚C at 15:00 in the middle of the steel web. Results of the three-dimensional FE computations show that the vertical temperature distribution in the steel web is almost uniform for the two selected days in June and December. This temperature gradient is expected given the high thermal conductivity of steel and its ability to adjust its temperature quickly to that of the surrounding environment. The computations show that on June 4, the maximum vertical temperature differential between the steel girder and the concrete deck occurred at time 18:00 and reached a maximum value of 11 ˚C (positive gradient), as shown in Figure 3(a), at Position I. This value is less than half that suggested by Kennedy and Soliman [8], which is 22.2 ˚C. The corresponding vertical temperature difference at Position II was only 6 ˚C. It is interesting to note that the absolute maximum temperature differential was reached at time 11:00 inside the concrete deck, and this is due to the low thermal diffusivity of concrete. This difference between the top surface and the mid-depth of the concrete deck reached a value of 11.2 ˚C. In fact, the top and bottom surfaces of the concrete deck are exposed to direct and ground reflected solar radiations at time 11:00 during the heating process. These radiations, in conjunction with the low thermal diffusivity of concrete, lead to a lower temperature at middepth of the deck as shown in Figure 3(a). After sunset, the thermal profile in the concrete deck, provided in Figure 3(a) at time 22:00, shows that the concrete is the warmest at mid-depth of the deck during the cooling process. This happens due to both top and bottom surfaces of the deck losing heat to the surrounding by convection faster than the concrete at mid-depth. On December 23, the vertical temperature distribution is almost uniform for the entire day due to low ambient temperature and solar radiation. The maximum negative differential temperature between the steel girder and the concrete deck reached a value of 2 ˚C at time 7:00 at Position I, as shown in Figure 3(b). This negative thermal gradient is due to the start of the heating process after sunrise. After sunset (after 17:00), the temperature distribution in the bridge reached almost a steady state with a maximum vertical temperature difference of 0.4 ˚C. Therefore, the cooling process had negligible effects on the thermal gradient in December under normal environmental conditions. 120 80 60 40 20 0 45 (a) Temperature (˚C) 10 15 (b) 40 35 20 25 30 35 40 11:00 18:00 22:00 30 25 20 15 Elevation (in) 100 Elevation (cm) Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. 4.3 Vertical Temperature Distribution -15 -10 -5 10 5 0 Figure 3. Vertical temperature distribution at Poisiton I (a) (June 4), (b) (7:00, December 23) © ASCE Congress on Technical Advancement 2017 9 The AASTO LRFD Bridge Design Specification [4] provides provisions for the vertical temperature gradient in composite bridges. AASHTO LRFD [4] recommends that the vertical temperature distribution be uniform in the steel girders and linear in the superimposed concrete deck. The temperature differential in the concrete deck is based on the map of solar radiation zones in the United States, in which the city of Fargo, ND is located in zone 2. For zone 2, the AASHTO LRFD Bridge Design Specification [4] recommends a maximum positive vertical temperature differential of 18.9 ˚C between the top and bottom surfaces of the concrete deck, with a linear profile in between. The AASHTO temperature gradient and that proposed by Kennedy and Soliman [8] are compared to that computed using the thermo-elastic analysis of the FE model in June, as shown in Figure 4. 120 45 (a) 25 30 35 40 45 50 30 25 60 FE Model 40 AASHTO 15 20 Kennedy & Soliman (1987) 10 0 Temperature (˚C) 20 5 0 40 100 35 20 45 (b) Elevation (in) 80 120 40 100 Elevation (cm) Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. 4.4 Comparison with Existing Models 80 35 20 25 30 35 40 50 30 25 60 20 15 40 10 20 0 45 5 Temperature (˚C) 0 Figure 4. Vertical temperature distribution at position I on June 4 at: (a) (18:00), (b) (11:00) Figure 4(a) shows that the shape of the obtained vertical temperature distribution is very similar to the one proposed by AASHTO and by Kennedy and Soliman [8]; however, the AASHTO specification provides a maximum temperature differential of at least 7.9 ˚C higher than the one obtained in this study at Position I (almost 13 oC at Position II). This difference is even more for the model proposed by Kennedy and Soliman [8], which reaches 11 oC at Position I and about 16oC at Position II. This very conservative approximation adopted by AASHTO leads to considerable error in assessing the thermal stresses in composite steel-concrete bridges in locations as the one considered here. Figure 4(b) shows a comparison of the vertical temperature distribution proposed by AASHTO and by Kennedy and Soliman [8] to that obtained in this study at time 11:00. The relative error between the two maximum temperature differentials in both profiles (AASHTO and proposed) is 7.7 ˚C. Another issue to consider is the nonlinearity of the thermal profile when compared with the linear distribution proposed by AASHTO. In fact, the vertical temperature distribution in the concrete deck was close to being linear only in the afternoon between time 14:00 and 19:00, and nonlinear for the remainder of the day. This nonlinearity in the temperature distribution will produce a nonlinear thermal strain component and its effects on the design of the bridge require further investigation. © ASCE Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. Congress on Technical Advancement 2017 10 It is thereefore eviden nt that the vertical thermaal gradient pproposed by AASHTO ffor the considdered climate is i conservative in the concrete c decck due to seeveral reasonns: a) AASHTO complletely ignores the t effect off the depth of o the concreete deck andd its effect oon the maxim mum temperrature differentiial. b) The thermal pro ofile prescriibed by AA ASHTO is bbased on a one dimenssional conductio on analysis,, and thus the t solar rad diation refleected from the ground is ignored. This reflected radiation deecrease the positive p therrmal gradiennt through hheating the bbottom surfaace of the conccrete deck. c) The AASHTO mod del does nott take the ooverhang-to--depth ratioo into consideraation. This ratio r has an effect on th he maximum m temperatuure differentiial of the crritical exterior beam b throug gh the shadin ng effect. d) A linear verrtical temperrature distribbution is assuumed by AASH HTO in the concrete deeck at all tim mes. e) AAS SHTO does not accountt for the effeect of pre-existing transverse construction cracks in n the concreete deck, whhose effect teends to lesseen the average temperature t differential between thee concrete deeck and the ssteel beam. The AASHTO LRFD L [4] seems s to prrovide a sim mple and ggeneral therm mal gradiennt for compositte bridges for f various geographic regions. Hoowever, it iis shown heere to be ooverly conservaative and sim mplistic. By incorporatin ng the aforeementioned points, the A AASHTO m model can be revised r to obtain a morre accurate thermal graadient and tthe correspoonding maxiimum temperatu ure differen ntials. This accuracy a willl help desiggners achievve a more opptimal desiggn by better esttimating and d accounting for the effeccts of the theermal stressees in compossite bridges. 4.5 Effeccts of Transvverse Deck Cracks C The com mputations performed p show s that the t deck traansverse craacks have aan apparentt and significan nt effect on the t thermal profile of th he compositee bridge. Thiis effect is m more noticeabble in the summ mer than it iss in the winter. Yet, a veery high win d speed duriing the winteer could leadd to a negative gradient at the crack po osition due to t air flow tthrough the cracks. Thiss effect mannifests itself in a reduction of o the positiive thermal gradient g betw ween the topp and bottom m of the conncrete deck, an nd can be demonstrated d d by plotting g the therm mal profiles at Positionss I and II oor by examinin ng the temp perature con ntours for th he positive aand negativve thermal ggradient casses in Figure 5.. The variatio on in the tem mperature diffferentials between Posiitions I and III ranges bettween 3.1 and 5.6 ˚C in Ju une. The hig ghest differeence of 5.6 ˚C is reachhed at time 21:00 durinng the p of th he bridge. In n fact, the bridge b will eexperience fa faster coolingg near the ccracks cooling process due to the faster heatt loss by convection at th he crack surfface. (a)) (b) Figure F 5. Teemperature distribution d at a (a) 20:00,, June 4, (b) 10:00, Deceember 23 © ASCE Congress on Technical Advancement 2017 Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. The vertical temperature distribution in the concrete deck at Position I was nonlinear for most of the day, while that at Position II has a much lower degree of nonlinearity due to the crack’s effect in helping the concrete to quickly adjust its internal temperature to the ambient conditions. Various codes and researchers ignore the temperature differentials in the longitudinal direction of the bridge and assume them to be negligible when compared with the vertical temperature distribution. The difference in the longitudinal temperature distribution is said not to exceed 15% of the vertical temperature differential suggested by AASHTO [4]. The computations presented here however indicate a maximum temperature differential of 6.3 ˚C in the longitudinal direction of the bridge when compared to the vertical temperature differential of 9.4 ˚C. The longitudinal temperature differential thus constitutes more than 67 % of the vertical temperature differential, and the effect of this nonlinear longitudinal thermal gradient on the development of cracks and nonlinear strains requires further assessment and cannot be treated in a trivial manner. 5. Conclusions and Recommendations The effect of pre-existing transverse construction cracks in the concrete deck on the thermal gradient in composite steel-concrete bridges was investigated using a non-linear transient thermal analysis in 3D FE formulation. The following main conclusions are reached: 1) The maximum positive vertical temperature differential between the concrete deck and steel girder reached a maximum value of 11 ˚C in the summer, while vertical thermal gradient is almost uniform for the entire day in winter, under normal environmental conditions. 2) For composite bridges in geographic regions having climates similar to the one considered in this case study, the AASHTO LRFD Bridge Design Specification is overly conservatively and overestimates the vertical thermal gradient, which leads to significant error in assessing the thermal stresses in composite steel-concrete bridges. The current AASHTO model is based on the 1989 guide specifications, and needs to be updated with new findings taken into considerations. 3) The FE model analysis show a nonlinear vertical temperature distribution in the concrete deck when compared to the linear distribution proposed by AASHTO and other previously-suggested models. This nonlinearity will create a nonlinear strain component that will create stresses currently not considered by designers. 4) The longitudinal thermal gradient is more than 67% of the vertical one. Its impact on the self-equilibrating transverse stresses in the bridge due to sectional continuity, Poisson’s effect, and displacement compatibility at the interface of consecutive sections needs further assessment. While the conclusions arrived at in this study can be generally applicable to similar bridges in somewhat similar climates, they remain specific to the bridge case under investigation. More numerical and experimental studies should be conducted on bridges under different weather conditions to draw a better assessment of the accuracy of currently used thermal gradient models, such as that of AASHTO in more moderate climates. References [1] [2] © ASCE Zuk, W. (1961). “Thermal and shrinkage stresses in composite beams.” ACI Journal, 58(9), 1529-1558. Imbsen, R. A., Vandershaf, D. E., Schamber, R. A., and Nutt, R. V. (1985). “Thermal effects in concrete bridge superstructures.” (NCHRP Report No. 276). Washington, DC: Transportation Research Board. 11 Congress on Technical Advancement 2017 [3] Downloaded from ascelibrary.org by RMIT UNIVERSITY LIBRARY on 01/03/19. Copyright ASCE. For personal use only; all rights reserved. [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] © ASCE Fu, H. C., Ng, S. F., and Cheung, M. S. (1990). “Thermal behavior of composite bridges.” J. Struct. Eng., 116(12), 3302-3323. AASHTO . (2012). “LRFD Bridge Design Specifications.” Washington, DC. Ramey, G. E., and Wright, R. (1994). “Assessing and enhancing the durability/longevity performances of highway bridges.” (HRC Research Project 2-13506). Alabama: Auburn University. Berwanger, C. (1983). “Transient thermal behavior of composite bridges.” J. of Struct. Eng., 109(10), 2325-2339. AASHTO (Association of State Highway and Transportation Officials). (1989). “Guide Specifications, Thermal Effects in Concrete Bridge Superstructures.” Washington, DC. Kennedy, J. B., and Soliman, M. H. (1987). “Temperature distribution in composite bridges.” J. of Struct. Eng., 113(3), 475-482. Chen, Q. (2008). “Effects of thermal loads on Texas steel bridges.” Doctoral dissertation. Emanuel, J. H., and Taylor, C. M. (1985). “Length-Thermal Stress Relations for Composite Bridges.” J. of Struct. Eng., 111(4), 788-804. Tong, M., Tham, L. G., and Au F. (2002). “Extreme thermal loading on steel bridges in tropical region.” J. of Bridge Eng., 7(6), 357-366. Ramey, G. E., Wolff, A. R., and Wright, R. L. (1997). “Structural design actions to mitigate bridge deck cracking.” Practice Periodical on Structural Design and Construction, 2(3): 118-124. PCA (Portland Cement Association). (1970). “Final report – Durability of concrete bridge decks – A cooperative study.” State Highway Departments of California, Illinois, Kansas, Michigan, Minnesota, Missouri, New Jersey, Ohio, Texas, and Virginia. Cheng, T. T. H., and Johnston, D. W. (1985). “Incidence assessment of transverse cracking in concrete bridge decks: Construction and material considerations.” (Report No. FHWA/NC/85-002,1). Kosel, H. C., and Michols, K. A. (1985). “Evaluation of concrete deck cracking for selected bridge deck structures of Ohio Turnpike.” Ohio: Construction Technology Laboratory, Ohio Department of Transportation. Krauss, P. D., and Rogalla, E. A. (1996). “Transverse cracking in newly constructed bridge decks.” Transportation Research Board, (NCHRP Report No. 380). Washington, DC. Emanuel, J. H., and Hulsey, J. L. (1978). “Temperature distributions in composite bridges.” Journal of the Structural Division, 104(1), 65-78. Ibrahim, A. M. (1995). “Three-dimensional thermal analysis of curved concrete box girder bridges.” Master’s thesis, Concordia University, Montreal. ASHRAE (American Society of Heating and Air conditioning Engineers). (1959). “Heating, ventilating, air conditioning guide.” 37(52). Duffie, J. A., and Beckman, W. A. (1991). Solar engineering of thermal process, Wiley, New York. NREL. (n.d.). National solar radiation database. Elbadry, M. M., and Ghali, A. (1983). “Temperature variations in concrete bridges.” J. of Struct. Eng., 109(10), 2355-2374. Moorty, S., and Roeder, C. W. (1992). “Temperature-dependent bridge movements.” J. Struct. Eng., 118(4), 1090-1105. 12