Wind-Tunnel Development and Trends In
Content
Journal of Wind Engineering and Industrial Aerodynamics 91 (2003) 355–370
Wind-tunnel development and trends in applications to civil engineering
Jack E. Cermaka,b
b a Department of Civil Engineering, Colorado State University, Fort Collins, CO 80523, USA Cermak Peterka Petersen, Inc. (CPP, Inc.), Wind Engineering Consultants, Fort Collins, CO 80524, USA
Received 20 November 2001; received in revised form 12 September 2002; accepted 13 September 2002
Abstract A review is presented on wind tunnels capable of simulating natural winds, the boundarylayer wind tunnel (BLWT), and trends in their extensive use in civil-engineering practice. BLWTs and data-acquisition systems, as they evolved to meet needs in civil engineering, are described. Advancements are highlighted for the types of wind-load information now available to structural engineers and architects by BLWT tests using the advanced data-acquisition systems–the high-frequency base balance (H-FBB) and the synchronous multi-pressure sensing system (SM-PSS). Trends in applications of BLWT tests to determine wind effects on structures and to investigate wind-related environmental problems are described. Technical details of the BLWT, the H-FBB, the SM-PSS, and their proper use are available in references cited and are not restated in this review. r 2002 Elsevier Science Ltd. All rights reserved.
Keywords: Boundary-layer wind tunnels; High-frequency base balance; Synchronous multi-pressure systems; Wind-load tests; Wind environment tests
1. Introduction Within the last 30 years wind tunnels have evolved as an indispensable aid to the practice of civil engineering. Major applications to civil engineering are through wind-tunnel tests for wind effects on structures. Another important class of civilengineering applications in which wind-tunnel investigations contribute valuable information is treatment of wind-related environmental problems. Wind tunnels and trends in their applications to civil-engineering practice are reviewed herein.
E-mail address: [email protected] (J.E. Cermak). 0167-6105/03/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 1 6 7 - 6 1 0 5 ( 0 2 ) 0 0 3 9 6 - 3
356
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
Emphasis is on advancements in the types of information that have become accessible due to recent data-acquisition system developments rather than on technical details of test procedures. Information on requirements for boundary-layer wind simulation, model construction, and procedures for various types of tests is available in Refs. [1–5]. Details presented in Ref. [4] are particularly comprehensive. In this review, advancements in techniques for creating air flow to obtain wind loads on objects by measurements on small-scale models are summarized. Efforts in this regard were reported in 1742 [6] and advanced to development of the boundarylayer wind tunnel (BLWT) concept in 1958 [7] which is in general use today. The benefits of BLWT tests on structures were not fully realized until new concepts and capabilities for data-acquisition systems were developed. Two systems are described that have greatly expanded the scope of information available to the practicing engineer—(1) the high-frequency base balance (H-FBB) reported in 1975 [8], and (2) the synchronous multi-pressure sensing system (SM-PSS) reported in the English literature in 1991 [9]. Wind-tunnel investigations for wind-engineering applications are discussed in a variety of publications that include Refs. [1–10]. In the following sections, applications that are indicative of current trends are emphasized.
2. Development of the boundary-layer wind tunnel Attempts to determine wind forces on objects using small-scale models in air flow generated by a variety of methods have been reported in the literature. The earliest accounts were published in 1742 [6] and 1759 [11] where flow was realized by rotating the object on an arm through still air. Approximately 150 years later, Australian studies on wind loads for houses were reported [12] in which models placed on a ground plane were subjected to an open air jet. At about the same time, 1894–1895, pressure studies were conducted in Denmark [13] on simple building models. Flow in this case was induced through a test section (1.02 m in length, 0.23 Â 0.11 m in cross section) installed in the wall of a gas-works smoke stack. None of the foregoing flows simulated features of natural winds in the atmospheric boundary layer. During the next 50 years, 1900–1950, pressure measurements on model buildings and mass-transport studies were made in various wind tunnels. However, wind tunnels of this period were not designed to simulate natural winds [14]. Wind-tunnel studies at the National Physical Laboratory [15] and at Colorado State University [16] revealed that physical modeling of wind effects requires a properly simulated boundary-layer flow. This finding was reinforced by comparison of mean pressure measurements at the Wind Laboratory, Technical University of Denmark, on a 1:20 scale model in a wind tunnel with field measurements on the full-scale building [17]. A wind tunnel for simulation of the atmospheric boundary layer was designed during 1955–1957 [7] and constructed in the period 1960–1962 [18]. Plan and elevation drawings of the original boundary-layer wind tunnel are shown in Fig. 1. Uniqueness of this facility is derived from the long test section which allows for
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
357
Fig. 1. BLWT, Colorado State University [7,19].
Fig. 2. The University of Western Ontario BLWT II [20].
development of thick turbulent boundary layers, heating and cooling capabilities for creating thermally stratified boundary layers and a flexible ceiling for adjustment of pressure along the test-section length. Details of this BLWT are given in [2–4,18]. Several major boundary-layer wind tunnels with advanced features were designed and constructed between 1980 and 1995. The most noteworthy are facilities at the University of Western Ontario (Fig. 2), Monash University (Fig. 3) and the Public Works Research Institute in Tsukubu, Japan (Fig. 4).
358
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
Fig. 3. The Monash University BLWT (vertical section)—courtesy Professor W.H. Melbourne (1994).
Fig. 4. The Public Works Research Institute BLWT, Tsukuba, Japan [21].
Advanced features of the BLWT shown in Fig. 2 are a water channel in the return flow section for studies of wind and wave forces on offshore structures and an automated surface roughness system in the high-speed test section. The BLWT shown in Fig. 3 incorporates three test sections designed to accommodate different
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
359
types of testing. Tests on civil and some aeronautical structures can be made in the first test section downstream from the fans at higher Reynolds numbers than in the BLWTs of Figs. 1 and 2. In the next test section noise generated by flow over motor vehicles can be studied. The third test section provides heating of the ceiling to create stably stratified boundary layers for environmental studies of dispersion. Design of the BLWT of Fig. 4 provides an advanced capability for testing long-span bridges. The 41-m wide test section can accommodate aeroelastic tests on full-bridge models at favorable length scales. The closed-circuit BLWTs shown in Figs. 1–4 provide flow free from external disturbances and laboratory space free from disturbing air currents. However, opencircuit BLWTs similar to the facility shown in Fig. 5 have been constructed throughout the world since about 1965 and more are under construction. Major advantages of the open-circuit BLWT are lower construction costs and a requirement of less laboratory space than for a closed-circuit BLWT with the same test-section length. Studies [22,23] have demonstrated that by placing flow-adjustment devices such as spires, vortex generators, and fences at the test-section entrance acceptable turbulent boundary layers can be realized in test sections shorter than for the BLWTs shown in Figs. 1–5. Furthermore, these devices can be used to thicken boundary layers in the long test-section BLWTs [24]. Characteristics of 241 low-speed wind-tunnels, as of about 1984, in the United States, Australia, Belgium, Canada, France, India, Italy, Japan, Netherlands, New Zealand, Spain, Sweden, Switzerland, United Kingdom, and West Germany have been published [25]. The list includes both aeronautical and boundary-layer wind tunnels. A key difference between the two types of facilities is test-section length.
Fig. 5. Open-circuit BLWT, Colorado State University [19].
360
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
3. Advancements in data-acquisition systems Modern capabilities of testing for civil-engineering applications (as well as architectural and urban-planning applications) have been made possible by development of advanced data-acquisition and processing systems [26]. The following systems have revolutionized the scope and detail of test results for purposes of design and analysis: (1) the high-frequency base balance (H-FBB), and (2) the synchronous multi-pressure scanning system (SM-PSS). Brief descriptions of both measurement systems are presented to provide a basis for discussion of their applications. 3.1. H-FBB characteristics Basic features of typical H-FBBs in use today are illustrated in Fig. 6. Development of this device began in the 1970s [27,8] and progressed as more sensitive strain-measurement arrangements became available [28]. With sufficiently low-mass models of high-stiffness and high-stiffness balances, natural frequencies of model/balance systems are nominally at least 100 Hz for moments about the three orthogonal axes (x, y, and z). Simultaneous measurements of the x, y, and z base moments and x and y base shears provide data for a variety of applications that include determination of dynamic responses. During the last 15 years H-FBB test results have become widely used by structural engineers and have replaced use of the costly aeroelastic test for many structures. However, the static H-FBB test does not account for negative aerodynamic damping that may occur for high wind speeds or structures that experience significant lateral deflections. This deficiency can cause wind loads predicted by a H-FBB test to be too low for some structures [29]. A major advantage of the H-FBB test is that wind loads (the aerodynamic admittance) obtained from a single test can be used to calculate
Fig. 6. High-frequency base balance (Courtesy Cermak Peterka Petersen, Inc.).
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
361
structural responses for a range of structural properties provided the exterior structural geometry is not altered. 3.2. SM-PSS characteristics A long-time aspiration has been the ability to acquire instantaneous distributions of pressures continuously on all exterior surfaces of a structure. Within the last 10 years development of SM-PSSs has made this possible. Initial development took place in Japan during 1986 [30] with progress reported in 1991 [9] and 1994 [31]. Aggregate sampling rates vary from about 100,000–500,000 samples per second and the number of pressure taps sampled simultaneously may vary from 500 to 1000 for systems at different laboratories. For a structure with 500 pressure taps, a 100,000 aggregate sampling rate SM-PSS will yield 200 pressure distributions each second.
4. Trends in applications of wind-tunnel testing to civil engineering Civil engineering and architectural applications of wind-tunnel testing may be divided into two broad classes—(1) wind effects on structures, and (2) environmental effects. Trends in types of wind effects on structures available to the engineer and architect brought about by development of the H-FBB and SM-PSS are described. Wind-related environmental problems associated with civil-engineering practice are noted that are frequently studied by physical modeling in BLWTs. Section 4 describes the types of information that may be derived from properly conducted BLWT tests. Test conditions that must be satisfied to obtain reliable wind-effect data are not presented herein. They are available in Ref. [4] with detailed discussions. 4.1. BLWT studies of wind effects on structures Structures commonly subjected to BLWT tests may be categorized according to the expected significance motion of the structure will have on increasing dynamic excitation—a phenomenon referred to as negative aerodynamic damping (NAD). Structures expected to have insignificant and significant NAD are listed as follows: Insignificant NAD High-rise buildings (massive, broad, complex environment) Low-rise buildings (hp 18 m, n>1 Hz) Large-area roofs (n>1 Hz) Significant NAD High-rise buildings (lightweight, slender, exposed) Towers, tall chimneys, masts Long-span bridges Iced transmission lines
where n is the fundamental natural frequency and h is building height. Advancements in applications of BLWT tests following development of the
362
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
H-FBB and SM-PSS have been primarily for structures in the Insignificant NAD category. The following commentary is focused on these advancements. 4.1.1. H-FBB tests The H-FBB test is used primarily to obtain wind loads on tall structures and buildings of medium-height. This test provides direct measurement of time series for base moments and shears (if needed) resulting from instantaneous overall wind loads. These data can be processed to provide a variety of information for design and analysis that include the following: A. Mean, rms, and peak overall wind loads. B. Correlations for components of base moments, base shears, and moment/shear combinations. C. Spectra of base loads. D. Approximate vertical distribution of peak loads when the H-FBB test is accompanied by a non-synchronous pressure test (a common occurrence). When two high-rise buildings are on a common foundation, connected by a sky bridge or closely spaced with potential for bumping, each building can be mounted on separate H-FBBs and base-load data acquired simultaneously [32]. Major benefits derived from development of the H-FBB may be summarized as follows: A. Statistics of overall wind loads became accessible. B. Tests can be performed quickly and for low cost compared to aeroelastic tests. C. For a given building geometry only one test is necessary to determine dynamic responses for various sets of structural properties and wind speeds. D. Test data (the aerodynamic admittance) can be acquired prior to determination of structural properties by the structural engineer. E. Findings of a H-FBB test can be used to determine the need for an aeroelastic test and the likely wind directions for critical responses. 4.1.2. SM-PSS tests By weighting (electronically) the instantaneous pressure at each pressure tap according to its tributary area and influence function; instantaneous forces and moments, their fluctuation statistics, and correlations can be determined. Useful applications are determination of wind loads and dynamic responses for high-rise buildings, low-rise buildings, modal areas of large span roofs, free-standing walls, canopies, various structural components, and on structures forming an interconnected group [31,33]. High-speed scanning systems, now available in the major wind-engineering laboratories, greatly extend the types of information that can be derived from BLWT tests for structural engineers and architects. Wind loads on high-rise buildings are sometimes determined by a SM-PSS test rather than a H-FBB test [34,35]. The SM-PSS test can provide all
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
363
information listed for a H-FBB test and, in addition, the following information and advantages: A. Accurate vertical distributions of peak shears and moments (including torsion) [34]. B. Design pressures for cladding are obtained from the same SM-PSS test data used to determine the wind loads. C. The dynamic analysis can be performed accurately for various mode shapes of the deflected building, rather than only a linear mode shape assumed for the H-FBB test, with approximate corrections needed for nonlinear shapes [36]. D. Internal pressures resulting from breached cladding elements and their correlation with external pressures on unbreached cladding elements can be determined from time series of the simultaneous pressure data [37]. Use of the SM-PSS test may be questionable for some high-rise buildings. If the building is very slender, there may be insufficient space in the model to accommodate pressure tubing for a large number of taps. Furthermore, if the building geometry is very complex, the number and location of taps required to derive the overall fluctuating wind loads are uncertain. In both of these circumstances a H-FBB test would be more appropriate. In the last decade, BLWT tests for wind effects on low-rise buildings have become very active [38]. This trend has been stimulated because SM-PSSs are available to obtain comprehensive wind-load data [39] and, in no small measure, because wind damage to low-rise buildings has become recognized as the cause of major economic loss. Many studies of wind loads on low-rise buildings using SM-PSSs have been made for the purpose of improving building codes and standards. Studies of wind pressure on roofs of various geometries [40–43], correlation of internal and areaaveraged external pressures [44], and the effects of surroundings on pressures for low-rise buildings [45] are examples of investigations contributing to building code development. Database-assisted design for wind loads [46] is in a state of development [47]. Time series of instantaneous pressure distributions for different wind directions are an essential component of the database. Measurements with SM-PSSs have made pressure (aerodynamic) databases accessible for simple low-rise buildings [48,49]. Exploratory studies of database-assisted design have demonstrated that such designs can be risk-consistent and may be more resistant to damage as well as more economical [50]. Use of BLWT tests with SM-PSS data acquisition for creation of pressure databases is expected to increase for some classes of building geometry. The advent of SM-PSSs has contributed greatly to design of large-area roof systems. Simultaneously sensed pressures at a large number of locations on modal areas can be processed to obtain time series of instantaneous modal wind loads and their spectra. With this information, dynamic responses of different modes can be computed [51] and combined with dynamic background (non-resonant) loading. Mean and fluctuating overall loads on parabolic domes determined by a SM-PSS have been reported [52].
364
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
4.1.3. Aeroelastic-model tests Aeroelastic-model tests are indispensable for design of civil-engineering structures that experience wind induced deformations and/or motions that in turn tend to increase the wind loads. Phenomena of this nature may result in aeroelastic instability with a possible catastrophic outcome or lesser extreme effects such as unacceptable accelerations and deformations. Generally, structures that are flexible, low mass, and lightly damped fall in this category of wind/structure interaction. Details and applications of aeroelastic-model tests are well documented for tall structures (buildings, towers, chimneys, masts), long structures (long-span bridges, transmission lines, cable-supported pipelines) and special structures (cable-supported roofs and curtainwalls, cooling towers, air-supported roofs, fabric roofs, sculptures) [53,3–5,10]. Through an aeroelastic-model test the level of damping needed to reduce dynamic responses to acceptable magnitudes can be determined for buildings with significant NAD. An encouraging trend resulting from this capability and satisfactory performance of various types of dampers [54] is that more and more structural engineers are including added damping as part of the design process for tall structures [4,10]. Aeroelastic-model tests for long-span structures primarily employ two types of models—section models and full-structure models. There has been significant progress in tests using each type of model for long-span bridges. Traditionally, section models with two-degrees-of-freedom (vertical and torsional motions) have been used to obtain some of the flutter derivatives needed for analytical studies. In order to obtain all of the flutter derivatives involved, measurement of lateral motion associated with fluctuating drag is required. This has been accomplished using a three-degree-of-freedom suspension system [55]. The system permits simultaneous measurements for all three degrees of freedom. Section-model tests of long-span bridge sections do not account for correlation of gust wind speeds over the bridge length; interaction between the long span(s), towers and approach structures; effects of local topography on wind characteristics; or close proximity to ground and/or water surfaces. The full-bridge aeroelastic model test can account for the effect of these factors on dynamic responses of the bridge. However, the limited width of wind-tunnel test sections shown in Figs. 1–3 give rise to concern about Reynolds-number effects because the model scale must be small to fit the test section. This concern has been addressed by construction of the 41-m wide BLWT shown in Fig. 4 [21]. 4.1.4. Wind-related environmental tests The frequency of BLWT tests to treat wind-related environmental aspects of importance to many civil-engineering projects is increasing rapidly. Environmental phenomena most commonly investigated for wind-engineering applications are the following: 1. Transport and deposition of snow by wind. 2. Mean and peak gust wind speeds at outdoor locations of heavy pedestrian use.
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
365
3. Topographic effects on wind characteristics. 4. Dispersion of air pollutants. 5. Wind climate in urban areas. Brief comments and some references on BLWT testing for each of the foregoing subjects follow. 4.1.4.1. Transport and deposition of snow. Wind-driven snow can result in snow loads on building roofs of unusual configuration for which building codes prescribe neither magnitudes nor locations. In many instances the depths and locations of roof-top accumulations of simulated snow on model buildings are determined by a BLWT test. Details of this type of test have been published for simple roof features [56,57]. Similar tests are being used to locate and evaluate snow fence performance for reduction of snow drifting on highways, railways, and structures [58,59]. A series of BLWT tests to reduce snow drifting on an approach road to the Denver International Airport [60] is one of several recent studies of this nature. 4.1.4.2. Pedestrian-level winds. Tall buildings and closely spaced buildings create strong winds at pedestrian-use areas. Measurement of mean and peak gust wind speeds at many locations within the domain of a site under development is now an integral component of most BLWT tests for wind loads [61]. Concerns about the wind environment in urban settings has stimulated interest in pedestrian-level winds [62–64]. A few cities including Boston, Toronto, New York, and San Francisco require BLWT test data on pedestrian-level winds with and without the new structures in place prior to approval of the proposed development [4,65]. The need to meet this requirement is determined on the basis of project building heights, the number and spacing of buildings, and the site surroundings. 4.1.4.3. Topographic effects on wind characteristics. Winds at locations near or in areas of complex topography can be drastically different from conventional boundary-layer winds. Determination of design winds at such locations poses a major problem for the design engineer. Frequently this problem is treated by conducting a BLWT test using a topographic model [66–68]. Model scales used for this purpose vary from 1:2,000 to 1:10,000. Wind characteristics measured at these scales are replicated at scales in the range of 1:200 to 1:500 commonly used in BLWT tests for structures. 4.1.4.4. Dispersion of air pollutants by wind. Dispersion of air pollutants is of concern to civil engineers, architects, and city planners in efforts to provide acceptable air quality in areas surrounding pollutant sources. Effluents from building vents, industrial stacks, and automobiles constitute sources that are most prevalent and troublesome. Physical modeling in BLWTs is the most effective method to investigate dispersion of these effluents [2,4,69]. Design and location of exhaust/intake systems to minimize recirculation of pollutants for hospital, university, and industrial laboratories using BLWT test data is now common
366
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
practice [70]. The effects of wind on distribution of automobile emissions in city canyons have been studied by physical modeling [71]. A concentration data bank developed for a line source in one block of an idealized city enables (by superpositions and summations) prediction of concentrations for proposed freeways and/or alternative traffic routings. Modeling criteria and procedures for BLWT studies of dispersion are presented in the literature [2,3,4,72]. 4.1.4.5. Urban wind climate. Wind climate in cities is of great concern during low wind speeds accompanied by an elevated temperature inversion. This condition results in development of high concentrations of automobile exhausts. Studies using BLWT tests reveal that city planning may lead to some features that can be helpful in reducing emission concentrations [71–74]. Exploratory studies indicate that buildingheight distributions, building spacing, and direction of the approach wind should be taken into consideration to minimize adverse effects of urbanization [75].
5. Conclusions A review of BLWTs and currently available data-acquisition systems reveals that BLWT tests continue to provide ever more comprehensive wind-load information for structural design. The stimulus for this trend has been development and widespread use of the H-FBB and the SM-PSS. While applications of BLWT test data in civil-engineering practice have been most common for wind effects on structures, applications to urban environmental problems of pedestrian-level winds and air quality are becoming equally common. Capabilities of BLWT tests to quantify wind effects on structures and the urban environment are stimulating close working relationships between structural engineers, architects, and city planners.
Acknowledgements Discussions with colleagues at CPP, Inc. served to provide a broad perspective for preparation of this review. Compilation of the manuscript was accomplished by Ms. Gloria Garza.
References
[1] C. Scruton, The use of wind tunnels in industrial aerodynamic research, Department of Scientific and Industrial Research, Report NPL/Aero/411, National Physical Laboratory, Teddington, England, 1960, 50pp. [2] J.E. Cermak, Application of fluid mechanics to wind engineering — A Freeman Scholar Lecture, J. Fluids Eng. ASME 97 (1975) 9–38. [3] T.A. Reinhold (ed.), Wind Tunnel Modeling for Civil Engineering Applications, Proceedings of the International Workshop on Wind Tunnel Modeling Criteria and Techniques in Civil Engineering Applications, Gaithersburg, Maryland, Cambridge University Press, London, England, 1982, 688pp.
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
367
[4] ASCE, Wind Tunnel Studies of Buildings and Structures, ASCE Manuals and Reports on Engineering Practice No. 67, ASCE, Reston, Virginia, 1999, 207pp. [5] J.D. Holmes, Wind Loading of Structures, Spon Press, New York, NY, 2001, 350pp. [6] B. Robins, New Principles of Gunnery, J. Nourse, London, 1742. [7] J.E. Cermak, Wind tunnel for the study of turbulence in the atmospheric surface layer, Final Report Air Force Cambridge Center, Air Research and Development Command, Bedford, MA (Contract AF 19(604)-1706), 1958, 36pp. [8] R.E. Whitbread, The measurement of non-steady wind forces on small-scale building models, in: Proceedings of the fourth International Conference on Wind Effects on Buildings and Structures, Heathrow, England, 1975, pp. 567–574. [9] A. Steckley, S.L. Gamble, M. Accardo, P.A. Irwin, G.D. Schuyler, The synchronous pressure acquisition network (SPAN), in: Proceedings of the ASCE Ninth Structures Congress, Indianapolis, Indiana, 1991, pp. 556–559. [10] E. Simiu, R.H. Scanlan, Wind Effects on Structures, 3rd Edition, Wiley, New York, NY, 1996, 688pp. [11] J. Smeaton, An experimental investigation concerning the natural powers of water and wind, Phil. Trans. R. Soc. London 51 (1759–1760) 100–174. [12] W.C. Kernot, Wind pressures, Australasian Builder and Contractors News 13 (1893) 194. [13] J.O.V. Irminger, Nogle forsog over trykforholdene paa planer og legemer pa a virkede of luftstromninger, IngeniNren (1894) and Engineering News and Engineers, 1895. [14] J.E. Cermak, Development of wind tunnels for physical modeling of the atmospheric boundary layer (ABL), in: State of the Art Volume, Ninth International Conference on Wind Engineering, New Delhi, India, 1995, pp. 1–25. [15] A. Bailey, N.D.G. Vincent, Wind pressures on buildings including effects of adjacent buildings, J. Civil Eng. 20 (8) (1943) 243–275. [16] J.E. Cermak, Application of modeling techniques to mass transfer and evaporation studies, Paper Presentation, Centennial of Engineering, Chicago, IL, 1952, 15pp. [17] M. Jensen, The model-law for phenomena in natural wind, IngeniNren 2 (4) (1958) 121–128. [18] E.J. Plate, J.E. Cermak, Micrometeorological wind tunnel facility, description and characteristics, Final Report US Army Electronic Research and Development Activity, Fort Huachuca, Arizona (Contract DA-36-0309SC-80371), 1963, 51pp. [19] J.E. Cermak, Wind tunnel design for physical modeling of atmospheric boundary layers, Special Volume on Advances in Fluid Mechanics, J. Eng. Mech., ASCE 107(EM3) (1981) 623–642. [20] A.G. Davenport, N. Isyumov, J.P.C. King, M. Novak, D. Surry, B.J. Vickery, BLWT II: The design and performance of a new boundary layer wind tunnel, in: Proceedings of the Fifth US National Conference on Wind Engineering, Texas Tech University, Lubbock, TX 1985, 2A35—39. [21] T. Miyata, H. Yamada, K. Yokoyama, T. Kanazaki, T. Iijima, M. Tatsumi, Construction of boundary layer wind tunnel for long-span bridges, J. Wind Eng. Ind. Aerodyn. 42 (1992) 885–896. [22] P.A. Irwin, The design of spires for wind simulation, J. Wind Eng. Ind. Aerodyn. 7 (1981) 361–366. [23] N.J. Cook, Simulation techniques for short test-section wind tunnels: roughness, barrier and mixingdevice methods, in: Proceedings of the International Workshop on Wind Tunnel Modeling Criteria and Techniques in Civil Engineering Applications, Gaithersburg, Maryland, Cambridge University Press, London, England, 1982, pp. 126–136. [24] J.E. Cermak, Physical modeling of the atmospheric boundary layer (ABL) in long boundary layer wind tunnels (BLWT), in: Proceedings of the International Workshop on Wind Tunnel Modeling Criteria and Techniques in Civil Engineering Applications, Gaithersburg, Maryland, Cambridge University Press, London, England, 1982, pp. 97–125. [25] W.H. Rae, A. Pope, Low-Speed Wind Tunnel Testing, 2nd Edition, Wiley, New York, NY, 1984, 534pp. [26] J.E. Cermak, Progress in physical modeling for wind engineering, J. Wind Eng. Ind. Aerodyn. 54/55 (1995) 439–455. [27] J.E. Cermak, W.Z. Sadeh, G. Hsi, Fluctuating moments on tall buildings produced by wind loading, Nat. Bur. Stand. Building Sci. Ser. 30 (1970) 45–59.
368
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
[28] T. Tschanz, A.G. Davenport, The base balance technique for the determination of dynamic wind loads, J. Wind Eng. Ind. Aerodyn. 13 (1983) 429–439. [29] D.W. Boggs, Validation of the aerodynamc model method, J. Wind Eng. Ind. Aerodyn. 41–44 (1992) 1011–1022. [30] K. Fuji, K. Hibi, H. Ueda, Anew measuring system using electronically scanned pressure sensors (ESP) and some applications of the ESP system to a square building shape, 9th Symposium on Wind Engineering, Tokyo, 1986, pp. 313–318, (in Japanese). [31] J.A. Peterka, L.S. Cochran, D.W. Boggs, N. Hosoya, M. Downing, Simultaneous peak pressure measurements in the wind tunnel, in: Proceedings of the International Conference on Building Envelope Systems and Technology, Singapore, 1994, pp. 354–359. [32] J. Xie, P.A. Irwin, Application of the force balance technique to a building complex, J. Wind Eng. Ind. Aerodyn. 77&78 (1998) 579–590. [33] A. Steckley, M. Accardo, S.L. Gamble, P.A. Irwin, The use of integrated pressures to determine overall wind-induced response, J. Wind Eng. Ind. Aerodyn. 41–44 (1992) 1023–1034. [34] T.C.E. Ho, G.R. Lythe, N. Isyumov, Structural loads from the integration of simultaneously measured pressures, in: Proceedings of the International Conference on Wind Engineering, Copenhagen, Denmark, 1999. [35] P.A. Irwin, W.W. Kochanski, Measurement of structural wind loads using a high frequency pressure integration method, in: Proceedings of the ASCE Structures Congress XIII, Boston, MA, 1995, pp. 1631–1634. [36] D.W. Boggs, J.A. Peterka, Aerodynamic model tests of tall buildings, J. Eng. Mech. ASCE 115 (1989) 618–635. [37] L.S. Cochran, J.A. Peterka, On breached building envelopes and increased internal pressure, in: Proceedings of the International Conference on Building Envelopes Systems and Technologies, Ottawa, Canada, 2001, pp. 83–87. [38] Y. Uematsu, N. Isyumov, Wind pressures acting on low-rise buildings, J. Wind Eng. Ind. Aerodyn. 82 (1999) 1–25. [39] Y. Tamura, H. Kikuchi, K. Hibi, Wind load combinations and extreme pressure distributions on lowrise buildings, Wind Struct. 3 (4) (2000) 279–289. [40] T. Stathopoulos, Evaluation of wind loads on low buildings — a brief historical review, in: State of the Art Volume, Ninth International Conference on Wind Engineering New Delhi, India, 1995, pp. 155–179. [41] B. Bienkiewicz, Y. Sun, Local wind loading on the roof of a low-rise building, J. Wind Eng. Ind. Aerodyn. 45 (1992) 11–24. [42] Y. Uematsu, N. Isyumov, Peak gust pressures acting on the roof and wall edges of a low-rise building, J. Wind Eng. Ind. Aerodyn. 77&78 (1998) 217–231. [43] J.A. Peterka, N. Hosoya, S. Dodge, L. Cochran, J.E. Cermak, Area-averaged peak pressures in a gable roof vortex region, J. Wind Eng. Ind. Aerodyn. 77&78 (1998) 205–215. [44] F. Beste, J.E. Cermak, Correlation of internal and area-averaged wind pressures on low-rise buildings, in: Proceedings of the Third International Colloquium on Bluff Body Aero. and Applications, Blacksburg, VA, 1996. [45] D. Surry, J.X. Lin, The effect of surroundings and roof corner geometric modifications on roof pressures on low-rise buildings, J. Wind Eng. Ind. Aerodyn. 58 (1995) 113–138. [46] E. Simiu, J.H. Garrett, K. Reed, Development of computer-based model of standards and attendant knowledge-based and procedural systems, in: Proceedings of the Structure and Engineering in Natural Hazard Mitigation, ASCE, New York, 1993, pp. 841–846. [47] A. Rigato, P. Chang, E. Simiu, Database assisted design, standardization, and wind direction effects, J. Struct. Eng. ASCE 127 (8) (2001) 855–860. [48] J. Lin, D. Surry, Simultaneous time series of pressures on the envelope of two large low-rise buildings, BLWT-SS7-1997, Boundary-Layer Wind Tunnel Laboratory, The University of Western Ontario, London, Ontario, Canada, 1997. [49] P.C. Case, N. Isyumov, Wind loads on low buildings with 4: 12 gable roofs in open country and suburban exposures, J. Wind Eng. Ind. Aerodyn. 77&78 (1998) 107–118.
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
369
[50] T. Whalen, E. Simiu, G. Harris, J. Lin, D. Surry, The use of aerodynamic databases for the effective estimation of wind effects in main-force resisting systems: applications to low buildings, J. Wind Eng. Ind. Aerodyn. 77&78 (1998) 685–693. [51] D.W. Boggs, J.A. Peterka, Wind loads on arena roofs using aerodynamic models, in: Proceedings of the Structures Congress ’93, ASCE, Irvine, CA, 1993, pp. 514–519. [52] C.W. Letchford, P.P. Sarkar, Mean and fluctuating wind loads on rough and smooth parabolic domes, J. Wind Eng. Ind. Aerodyn. 88 (2000) 101–117. [53] C. Scruton, On the wind-excited oscillations of stacks, towers and masts, in: Proceedings of the Conference on Wind Effects on Buildings and Structures, National Physcial Laboratory, Teddington, Middlesex, England Symposium No. 16, 1965, pp. 798–836. [54] A. Kareem, T. Kijewski, Y. Tamura, Mitigation of motions of tall buildings with specific examples of recent applications, Wind Struct. 2 (3) (1999) 201–251. [55] L. Singh, N.P. Jones, R.H. Scanlan, O. Lorendeaux, Simultaneous identification of 3-DOF aeroelastic parameters, in: Proceedings of the Ninth International Conference on Wind Engineering, New Delhi, India, 1995, pp. 972–977. [56] N. Isyumov, M. Mikitiuk, Wind tunnel model tests of snow drifting on a two-level flat roof, J. Wind Eng. Ind. Aerodyn. 36 (1990) 893–904. [57] J. Wianecki, A. Chevallier, Excess loads on flat roofs of buildings with and without parapets under the action of wind and snow, in: Proceedings of the Second International Conference on Snow Engineering, Santa Barbara, CA, 1992, pp. 215–228. [58] N. Isyumov, M. Mikitiuk, P. Cookson, Wind tunnel modeling of snow with applications to snow fences, in: Proceedings of the First International Conference on Snow Engineering, Santa Barbara, CA, 1988, pp. 210–226. [59] J.D. Iversen, Small-scale modeling of snow-drift phenomena, in: Proceedings of the International Workshop on Wind-tunnel Modeling Criteria and Techniques in Civil Engineering Applications, Gaithersburg, Maryland, Cambridge University Press, London, England, 1982, pp. 522–545. [60] J.A. Peterka, Wind-tunnel tests — snow deposition and drifting on Denver International Airport Pena Boulevard, Final Report, Cermak Peterka Petersen, Inc., Fort Collins, CO, 1999, 156pp. [61] F.H. Durgin, Pedestrian level wind studies at the Wright Brothers facility, J. Wind Eng. Ind. Aerodyn. 44 (1992) 2253–2264. [62] M.J. Soligo, P.A. Irwin, C.J. Williams, G.D. Schuyler, Acomprehensive assessment of pedestrian comfort including thermal effects, J. Wind Eng. Ind. Aerodyn. 77&88 (1998) 753–766. [63] M.A. Ratcliff, J.A. Peterka, Comparison of pedestrian wind acceptability criteria, J. Wind Eng. Ind. Aerodyn. 36 (1990) 791–800. [64] R.M. Aynsley, W. Melbourne, B.J. Vickery, Architectural Aerodynamics, Applied Science Publishers Ltd., London, England, 1977, 254pp. [65] B.R. White, Analysis and wind-tunnel simulation of pedestrian-level winds in San Francisco, J. Wind Eng. Ind. Aerodyn. 41–44 (1992) 2353–2364. [66] L.S. Cochran, J.A. Peterka, Modeling flow over complex terrain, in: Proceedings of the 2001 Structures Congress and Exposition, ASCE, New York, 2001. [67] J.E. Cermak, Physical modeling of flow and dispersion over complex terrain, J. Boundry Layer Met. 30 (1984) 261–292. [68] R.N. Meroney, Wind-tunnel simulations of flow over hills and complex terrain, J. Ind. Aerodyn. 5 (1980) 297–321. [69] R.P. Hosker, Jr., Flow and diffusion near obstacles, in: Atmospheric Science and Power Production, US Department of Energy, Available as DE84005177 (DOE/TIC-27601) from National Technical Information Service, US Department of Commerce, Springfield, VA 22161, 1984, pp. 241–326. [70] R.L. Petersen, B.C. Cochran, J.J. Carter, Specifying exhaust and intake systems, ASHRAE J. 44 (8) (2002) 30–37. [71] J.B. Wedding, D.J. Lombardi, J.E. Cermak, Awind tunnel study of gaseous pollutants in city street canyons, J. Air Pollut. Control Assoc. 27 (1977) 557–566. [72] J.E. Cermak, K. Takeda, Physical modeling of urban air-pollutant transport, J. Wind Eng. Ind. Aerodyn. 21 (1985) 51–67.
370
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
[73] J.E. Cermak, A.G. Davenport, E.J. Plate, D.X. Viegas (eds.), Proceedings of the NATO Advanced Study Institute on Wind Climate in Cities, Waldbronn, Germany (1993), Kluwer Academic Press, Dordeckt, The Netherlands, 1995, 772pp. [74] P. Kastner-Klein, E. Fedorovich, M.W. Rotach, A wind-tunnel study of organized and turbulent air motions in urban street canyons, J. Wind Eng. Ind. Aerodyn. 89 (2001) 849–861. [75] H.E. Landsberg, The Urban Climate, Academic Press, Inc., New York, NY, 1981, 275pp.
Wind-tunnel development and trends in applications to civil engineering
Jack E. Cermaka,b
b a Department of Civil Engineering, Colorado State University, Fort Collins, CO 80523, USA Cermak Peterka Petersen, Inc. (CPP, Inc.), Wind Engineering Consultants, Fort Collins, CO 80524, USA
Received 20 November 2001; received in revised form 12 September 2002; accepted 13 September 2002
Abstract A review is presented on wind tunnels capable of simulating natural winds, the boundarylayer wind tunnel (BLWT), and trends in their extensive use in civil-engineering practice. BLWTs and data-acquisition systems, as they evolved to meet needs in civil engineering, are described. Advancements are highlighted for the types of wind-load information now available to structural engineers and architects by BLWT tests using the advanced data-acquisition systems–the high-frequency base balance (H-FBB) and the synchronous multi-pressure sensing system (SM-PSS). Trends in applications of BLWT tests to determine wind effects on structures and to investigate wind-related environmental problems are described. Technical details of the BLWT, the H-FBB, the SM-PSS, and their proper use are available in references cited and are not restated in this review. r 2002 Elsevier Science Ltd. All rights reserved.
Keywords: Boundary-layer wind tunnels; High-frequency base balance; Synchronous multi-pressure systems; Wind-load tests; Wind environment tests
1. Introduction Within the last 30 years wind tunnels have evolved as an indispensable aid to the practice of civil engineering. Major applications to civil engineering are through wind-tunnel tests for wind effects on structures. Another important class of civilengineering applications in which wind-tunnel investigations contribute valuable information is treatment of wind-related environmental problems. Wind tunnels and trends in their applications to civil-engineering practice are reviewed herein.
E-mail address: [email protected] (J.E. Cermak). 0167-6105/03/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 1 6 7 - 6 1 0 5 ( 0 2 ) 0 0 3 9 6 - 3
356
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
Emphasis is on advancements in the types of information that have become accessible due to recent data-acquisition system developments rather than on technical details of test procedures. Information on requirements for boundary-layer wind simulation, model construction, and procedures for various types of tests is available in Refs. [1–5]. Details presented in Ref. [4] are particularly comprehensive. In this review, advancements in techniques for creating air flow to obtain wind loads on objects by measurements on small-scale models are summarized. Efforts in this regard were reported in 1742 [6] and advanced to development of the boundarylayer wind tunnel (BLWT) concept in 1958 [7] which is in general use today. The benefits of BLWT tests on structures were not fully realized until new concepts and capabilities for data-acquisition systems were developed. Two systems are described that have greatly expanded the scope of information available to the practicing engineer—(1) the high-frequency base balance (H-FBB) reported in 1975 [8], and (2) the synchronous multi-pressure sensing system (SM-PSS) reported in the English literature in 1991 [9]. Wind-tunnel investigations for wind-engineering applications are discussed in a variety of publications that include Refs. [1–10]. In the following sections, applications that are indicative of current trends are emphasized.
2. Development of the boundary-layer wind tunnel Attempts to determine wind forces on objects using small-scale models in air flow generated by a variety of methods have been reported in the literature. The earliest accounts were published in 1742 [6] and 1759 [11] where flow was realized by rotating the object on an arm through still air. Approximately 150 years later, Australian studies on wind loads for houses were reported [12] in which models placed on a ground plane were subjected to an open air jet. At about the same time, 1894–1895, pressure studies were conducted in Denmark [13] on simple building models. Flow in this case was induced through a test section (1.02 m in length, 0.23 Â 0.11 m in cross section) installed in the wall of a gas-works smoke stack. None of the foregoing flows simulated features of natural winds in the atmospheric boundary layer. During the next 50 years, 1900–1950, pressure measurements on model buildings and mass-transport studies were made in various wind tunnels. However, wind tunnels of this period were not designed to simulate natural winds [14]. Wind-tunnel studies at the National Physical Laboratory [15] and at Colorado State University [16] revealed that physical modeling of wind effects requires a properly simulated boundary-layer flow. This finding was reinforced by comparison of mean pressure measurements at the Wind Laboratory, Technical University of Denmark, on a 1:20 scale model in a wind tunnel with field measurements on the full-scale building [17]. A wind tunnel for simulation of the atmospheric boundary layer was designed during 1955–1957 [7] and constructed in the period 1960–1962 [18]. Plan and elevation drawings of the original boundary-layer wind tunnel are shown in Fig. 1. Uniqueness of this facility is derived from the long test section which allows for
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
357
Fig. 1. BLWT, Colorado State University [7,19].
Fig. 2. The University of Western Ontario BLWT II [20].
development of thick turbulent boundary layers, heating and cooling capabilities for creating thermally stratified boundary layers and a flexible ceiling for adjustment of pressure along the test-section length. Details of this BLWT are given in [2–4,18]. Several major boundary-layer wind tunnels with advanced features were designed and constructed between 1980 and 1995. The most noteworthy are facilities at the University of Western Ontario (Fig. 2), Monash University (Fig. 3) and the Public Works Research Institute in Tsukubu, Japan (Fig. 4).
358
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
Fig. 3. The Monash University BLWT (vertical section)—courtesy Professor W.H. Melbourne (1994).
Fig. 4. The Public Works Research Institute BLWT, Tsukuba, Japan [21].
Advanced features of the BLWT shown in Fig. 2 are a water channel in the return flow section for studies of wind and wave forces on offshore structures and an automated surface roughness system in the high-speed test section. The BLWT shown in Fig. 3 incorporates three test sections designed to accommodate different
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
359
types of testing. Tests on civil and some aeronautical structures can be made in the first test section downstream from the fans at higher Reynolds numbers than in the BLWTs of Figs. 1 and 2. In the next test section noise generated by flow over motor vehicles can be studied. The third test section provides heating of the ceiling to create stably stratified boundary layers for environmental studies of dispersion. Design of the BLWT of Fig. 4 provides an advanced capability for testing long-span bridges. The 41-m wide test section can accommodate aeroelastic tests on full-bridge models at favorable length scales. The closed-circuit BLWTs shown in Figs. 1–4 provide flow free from external disturbances and laboratory space free from disturbing air currents. However, opencircuit BLWTs similar to the facility shown in Fig. 5 have been constructed throughout the world since about 1965 and more are under construction. Major advantages of the open-circuit BLWT are lower construction costs and a requirement of less laboratory space than for a closed-circuit BLWT with the same test-section length. Studies [22,23] have demonstrated that by placing flow-adjustment devices such as spires, vortex generators, and fences at the test-section entrance acceptable turbulent boundary layers can be realized in test sections shorter than for the BLWTs shown in Figs. 1–5. Furthermore, these devices can be used to thicken boundary layers in the long test-section BLWTs [24]. Characteristics of 241 low-speed wind-tunnels, as of about 1984, in the United States, Australia, Belgium, Canada, France, India, Italy, Japan, Netherlands, New Zealand, Spain, Sweden, Switzerland, United Kingdom, and West Germany have been published [25]. The list includes both aeronautical and boundary-layer wind tunnels. A key difference between the two types of facilities is test-section length.
Fig. 5. Open-circuit BLWT, Colorado State University [19].
360
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
3. Advancements in data-acquisition systems Modern capabilities of testing for civil-engineering applications (as well as architectural and urban-planning applications) have been made possible by development of advanced data-acquisition and processing systems [26]. The following systems have revolutionized the scope and detail of test results for purposes of design and analysis: (1) the high-frequency base balance (H-FBB), and (2) the synchronous multi-pressure scanning system (SM-PSS). Brief descriptions of both measurement systems are presented to provide a basis for discussion of their applications. 3.1. H-FBB characteristics Basic features of typical H-FBBs in use today are illustrated in Fig. 6. Development of this device began in the 1970s [27,8] and progressed as more sensitive strain-measurement arrangements became available [28]. With sufficiently low-mass models of high-stiffness and high-stiffness balances, natural frequencies of model/balance systems are nominally at least 100 Hz for moments about the three orthogonal axes (x, y, and z). Simultaneous measurements of the x, y, and z base moments and x and y base shears provide data for a variety of applications that include determination of dynamic responses. During the last 15 years H-FBB test results have become widely used by structural engineers and have replaced use of the costly aeroelastic test for many structures. However, the static H-FBB test does not account for negative aerodynamic damping that may occur for high wind speeds or structures that experience significant lateral deflections. This deficiency can cause wind loads predicted by a H-FBB test to be too low for some structures [29]. A major advantage of the H-FBB test is that wind loads (the aerodynamic admittance) obtained from a single test can be used to calculate
Fig. 6. High-frequency base balance (Courtesy Cermak Peterka Petersen, Inc.).
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
361
structural responses for a range of structural properties provided the exterior structural geometry is not altered. 3.2. SM-PSS characteristics A long-time aspiration has been the ability to acquire instantaneous distributions of pressures continuously on all exterior surfaces of a structure. Within the last 10 years development of SM-PSSs has made this possible. Initial development took place in Japan during 1986 [30] with progress reported in 1991 [9] and 1994 [31]. Aggregate sampling rates vary from about 100,000–500,000 samples per second and the number of pressure taps sampled simultaneously may vary from 500 to 1000 for systems at different laboratories. For a structure with 500 pressure taps, a 100,000 aggregate sampling rate SM-PSS will yield 200 pressure distributions each second.
4. Trends in applications of wind-tunnel testing to civil engineering Civil engineering and architectural applications of wind-tunnel testing may be divided into two broad classes—(1) wind effects on structures, and (2) environmental effects. Trends in types of wind effects on structures available to the engineer and architect brought about by development of the H-FBB and SM-PSS are described. Wind-related environmental problems associated with civil-engineering practice are noted that are frequently studied by physical modeling in BLWTs. Section 4 describes the types of information that may be derived from properly conducted BLWT tests. Test conditions that must be satisfied to obtain reliable wind-effect data are not presented herein. They are available in Ref. [4] with detailed discussions. 4.1. BLWT studies of wind effects on structures Structures commonly subjected to BLWT tests may be categorized according to the expected significance motion of the structure will have on increasing dynamic excitation—a phenomenon referred to as negative aerodynamic damping (NAD). Structures expected to have insignificant and significant NAD are listed as follows: Insignificant NAD High-rise buildings (massive, broad, complex environment) Low-rise buildings (hp 18 m, n>1 Hz) Large-area roofs (n>1 Hz) Significant NAD High-rise buildings (lightweight, slender, exposed) Towers, tall chimneys, masts Long-span bridges Iced transmission lines
where n is the fundamental natural frequency and h is building height. Advancements in applications of BLWT tests following development of the
362
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
H-FBB and SM-PSS have been primarily for structures in the Insignificant NAD category. The following commentary is focused on these advancements. 4.1.1. H-FBB tests The H-FBB test is used primarily to obtain wind loads on tall structures and buildings of medium-height. This test provides direct measurement of time series for base moments and shears (if needed) resulting from instantaneous overall wind loads. These data can be processed to provide a variety of information for design and analysis that include the following: A. Mean, rms, and peak overall wind loads. B. Correlations for components of base moments, base shears, and moment/shear combinations. C. Spectra of base loads. D. Approximate vertical distribution of peak loads when the H-FBB test is accompanied by a non-synchronous pressure test (a common occurrence). When two high-rise buildings are on a common foundation, connected by a sky bridge or closely spaced with potential for bumping, each building can be mounted on separate H-FBBs and base-load data acquired simultaneously [32]. Major benefits derived from development of the H-FBB may be summarized as follows: A. Statistics of overall wind loads became accessible. B. Tests can be performed quickly and for low cost compared to aeroelastic tests. C. For a given building geometry only one test is necessary to determine dynamic responses for various sets of structural properties and wind speeds. D. Test data (the aerodynamic admittance) can be acquired prior to determination of structural properties by the structural engineer. E. Findings of a H-FBB test can be used to determine the need for an aeroelastic test and the likely wind directions for critical responses. 4.1.2. SM-PSS tests By weighting (electronically) the instantaneous pressure at each pressure tap according to its tributary area and influence function; instantaneous forces and moments, their fluctuation statistics, and correlations can be determined. Useful applications are determination of wind loads and dynamic responses for high-rise buildings, low-rise buildings, modal areas of large span roofs, free-standing walls, canopies, various structural components, and on structures forming an interconnected group [31,33]. High-speed scanning systems, now available in the major wind-engineering laboratories, greatly extend the types of information that can be derived from BLWT tests for structural engineers and architects. Wind loads on high-rise buildings are sometimes determined by a SM-PSS test rather than a H-FBB test [34,35]. The SM-PSS test can provide all
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
363
information listed for a H-FBB test and, in addition, the following information and advantages: A. Accurate vertical distributions of peak shears and moments (including torsion) [34]. B. Design pressures for cladding are obtained from the same SM-PSS test data used to determine the wind loads. C. The dynamic analysis can be performed accurately for various mode shapes of the deflected building, rather than only a linear mode shape assumed for the H-FBB test, with approximate corrections needed for nonlinear shapes [36]. D. Internal pressures resulting from breached cladding elements and their correlation with external pressures on unbreached cladding elements can be determined from time series of the simultaneous pressure data [37]. Use of the SM-PSS test may be questionable for some high-rise buildings. If the building is very slender, there may be insufficient space in the model to accommodate pressure tubing for a large number of taps. Furthermore, if the building geometry is very complex, the number and location of taps required to derive the overall fluctuating wind loads are uncertain. In both of these circumstances a H-FBB test would be more appropriate. In the last decade, BLWT tests for wind effects on low-rise buildings have become very active [38]. This trend has been stimulated because SM-PSSs are available to obtain comprehensive wind-load data [39] and, in no small measure, because wind damage to low-rise buildings has become recognized as the cause of major economic loss. Many studies of wind loads on low-rise buildings using SM-PSSs have been made for the purpose of improving building codes and standards. Studies of wind pressure on roofs of various geometries [40–43], correlation of internal and areaaveraged external pressures [44], and the effects of surroundings on pressures for low-rise buildings [45] are examples of investigations contributing to building code development. Database-assisted design for wind loads [46] is in a state of development [47]. Time series of instantaneous pressure distributions for different wind directions are an essential component of the database. Measurements with SM-PSSs have made pressure (aerodynamic) databases accessible for simple low-rise buildings [48,49]. Exploratory studies of database-assisted design have demonstrated that such designs can be risk-consistent and may be more resistant to damage as well as more economical [50]. Use of BLWT tests with SM-PSS data acquisition for creation of pressure databases is expected to increase for some classes of building geometry. The advent of SM-PSSs has contributed greatly to design of large-area roof systems. Simultaneously sensed pressures at a large number of locations on modal areas can be processed to obtain time series of instantaneous modal wind loads and their spectra. With this information, dynamic responses of different modes can be computed [51] and combined with dynamic background (non-resonant) loading. Mean and fluctuating overall loads on parabolic domes determined by a SM-PSS have been reported [52].
364
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
4.1.3. Aeroelastic-model tests Aeroelastic-model tests are indispensable for design of civil-engineering structures that experience wind induced deformations and/or motions that in turn tend to increase the wind loads. Phenomena of this nature may result in aeroelastic instability with a possible catastrophic outcome or lesser extreme effects such as unacceptable accelerations and deformations. Generally, structures that are flexible, low mass, and lightly damped fall in this category of wind/structure interaction. Details and applications of aeroelastic-model tests are well documented for tall structures (buildings, towers, chimneys, masts), long structures (long-span bridges, transmission lines, cable-supported pipelines) and special structures (cable-supported roofs and curtainwalls, cooling towers, air-supported roofs, fabric roofs, sculptures) [53,3–5,10]. Through an aeroelastic-model test the level of damping needed to reduce dynamic responses to acceptable magnitudes can be determined for buildings with significant NAD. An encouraging trend resulting from this capability and satisfactory performance of various types of dampers [54] is that more and more structural engineers are including added damping as part of the design process for tall structures [4,10]. Aeroelastic-model tests for long-span structures primarily employ two types of models—section models and full-structure models. There has been significant progress in tests using each type of model for long-span bridges. Traditionally, section models with two-degrees-of-freedom (vertical and torsional motions) have been used to obtain some of the flutter derivatives needed for analytical studies. In order to obtain all of the flutter derivatives involved, measurement of lateral motion associated with fluctuating drag is required. This has been accomplished using a three-degree-of-freedom suspension system [55]. The system permits simultaneous measurements for all three degrees of freedom. Section-model tests of long-span bridge sections do not account for correlation of gust wind speeds over the bridge length; interaction between the long span(s), towers and approach structures; effects of local topography on wind characteristics; or close proximity to ground and/or water surfaces. The full-bridge aeroelastic model test can account for the effect of these factors on dynamic responses of the bridge. However, the limited width of wind-tunnel test sections shown in Figs. 1–3 give rise to concern about Reynolds-number effects because the model scale must be small to fit the test section. This concern has been addressed by construction of the 41-m wide BLWT shown in Fig. 4 [21]. 4.1.4. Wind-related environmental tests The frequency of BLWT tests to treat wind-related environmental aspects of importance to many civil-engineering projects is increasing rapidly. Environmental phenomena most commonly investigated for wind-engineering applications are the following: 1. Transport and deposition of snow by wind. 2. Mean and peak gust wind speeds at outdoor locations of heavy pedestrian use.
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
365
3. Topographic effects on wind characteristics. 4. Dispersion of air pollutants. 5. Wind climate in urban areas. Brief comments and some references on BLWT testing for each of the foregoing subjects follow. 4.1.4.1. Transport and deposition of snow. Wind-driven snow can result in snow loads on building roofs of unusual configuration for which building codes prescribe neither magnitudes nor locations. In many instances the depths and locations of roof-top accumulations of simulated snow on model buildings are determined by a BLWT test. Details of this type of test have been published for simple roof features [56,57]. Similar tests are being used to locate and evaluate snow fence performance for reduction of snow drifting on highways, railways, and structures [58,59]. A series of BLWT tests to reduce snow drifting on an approach road to the Denver International Airport [60] is one of several recent studies of this nature. 4.1.4.2. Pedestrian-level winds. Tall buildings and closely spaced buildings create strong winds at pedestrian-use areas. Measurement of mean and peak gust wind speeds at many locations within the domain of a site under development is now an integral component of most BLWT tests for wind loads [61]. Concerns about the wind environment in urban settings has stimulated interest in pedestrian-level winds [62–64]. A few cities including Boston, Toronto, New York, and San Francisco require BLWT test data on pedestrian-level winds with and without the new structures in place prior to approval of the proposed development [4,65]. The need to meet this requirement is determined on the basis of project building heights, the number and spacing of buildings, and the site surroundings. 4.1.4.3. Topographic effects on wind characteristics. Winds at locations near or in areas of complex topography can be drastically different from conventional boundary-layer winds. Determination of design winds at such locations poses a major problem for the design engineer. Frequently this problem is treated by conducting a BLWT test using a topographic model [66–68]. Model scales used for this purpose vary from 1:2,000 to 1:10,000. Wind characteristics measured at these scales are replicated at scales in the range of 1:200 to 1:500 commonly used in BLWT tests for structures. 4.1.4.4. Dispersion of air pollutants by wind. Dispersion of air pollutants is of concern to civil engineers, architects, and city planners in efforts to provide acceptable air quality in areas surrounding pollutant sources. Effluents from building vents, industrial stacks, and automobiles constitute sources that are most prevalent and troublesome. Physical modeling in BLWTs is the most effective method to investigate dispersion of these effluents [2,4,69]. Design and location of exhaust/intake systems to minimize recirculation of pollutants for hospital, university, and industrial laboratories using BLWT test data is now common
366
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
practice [70]. The effects of wind on distribution of automobile emissions in city canyons have been studied by physical modeling [71]. A concentration data bank developed for a line source in one block of an idealized city enables (by superpositions and summations) prediction of concentrations for proposed freeways and/or alternative traffic routings. Modeling criteria and procedures for BLWT studies of dispersion are presented in the literature [2,3,4,72]. 4.1.4.5. Urban wind climate. Wind climate in cities is of great concern during low wind speeds accompanied by an elevated temperature inversion. This condition results in development of high concentrations of automobile exhausts. Studies using BLWT tests reveal that city planning may lead to some features that can be helpful in reducing emission concentrations [71–74]. Exploratory studies indicate that buildingheight distributions, building spacing, and direction of the approach wind should be taken into consideration to minimize adverse effects of urbanization [75].
5. Conclusions A review of BLWTs and currently available data-acquisition systems reveals that BLWT tests continue to provide ever more comprehensive wind-load information for structural design. The stimulus for this trend has been development and widespread use of the H-FBB and the SM-PSS. While applications of BLWT test data in civil-engineering practice have been most common for wind effects on structures, applications to urban environmental problems of pedestrian-level winds and air quality are becoming equally common. Capabilities of BLWT tests to quantify wind effects on structures and the urban environment are stimulating close working relationships between structural engineers, architects, and city planners.
Acknowledgements Discussions with colleagues at CPP, Inc. served to provide a broad perspective for preparation of this review. Compilation of the manuscript was accomplished by Ms. Gloria Garza.
References
[1] C. Scruton, The use of wind tunnels in industrial aerodynamic research, Department of Scientific and Industrial Research, Report NPL/Aero/411, National Physical Laboratory, Teddington, England, 1960, 50pp. [2] J.E. Cermak, Application of fluid mechanics to wind engineering — A Freeman Scholar Lecture, J. Fluids Eng. ASME 97 (1975) 9–38. [3] T.A. Reinhold (ed.), Wind Tunnel Modeling for Civil Engineering Applications, Proceedings of the International Workshop on Wind Tunnel Modeling Criteria and Techniques in Civil Engineering Applications, Gaithersburg, Maryland, Cambridge University Press, London, England, 1982, 688pp.
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
367
[4] ASCE, Wind Tunnel Studies of Buildings and Structures, ASCE Manuals and Reports on Engineering Practice No. 67, ASCE, Reston, Virginia, 1999, 207pp. [5] J.D. Holmes, Wind Loading of Structures, Spon Press, New York, NY, 2001, 350pp. [6] B. Robins, New Principles of Gunnery, J. Nourse, London, 1742. [7] J.E. Cermak, Wind tunnel for the study of turbulence in the atmospheric surface layer, Final Report Air Force Cambridge Center, Air Research and Development Command, Bedford, MA (Contract AF 19(604)-1706), 1958, 36pp. [8] R.E. Whitbread, The measurement of non-steady wind forces on small-scale building models, in: Proceedings of the fourth International Conference on Wind Effects on Buildings and Structures, Heathrow, England, 1975, pp. 567–574. [9] A. Steckley, S.L. Gamble, M. Accardo, P.A. Irwin, G.D. Schuyler, The synchronous pressure acquisition network (SPAN), in: Proceedings of the ASCE Ninth Structures Congress, Indianapolis, Indiana, 1991, pp. 556–559. [10] E. Simiu, R.H. Scanlan, Wind Effects on Structures, 3rd Edition, Wiley, New York, NY, 1996, 688pp. [11] J. Smeaton, An experimental investigation concerning the natural powers of water and wind, Phil. Trans. R. Soc. London 51 (1759–1760) 100–174. [12] W.C. Kernot, Wind pressures, Australasian Builder and Contractors News 13 (1893) 194. [13] J.O.V. Irminger, Nogle forsog over trykforholdene paa planer og legemer pa a virkede of luftstromninger, IngeniNren (1894) and Engineering News and Engineers, 1895. [14] J.E. Cermak, Development of wind tunnels for physical modeling of the atmospheric boundary layer (ABL), in: State of the Art Volume, Ninth International Conference on Wind Engineering, New Delhi, India, 1995, pp. 1–25. [15] A. Bailey, N.D.G. Vincent, Wind pressures on buildings including effects of adjacent buildings, J. Civil Eng. 20 (8) (1943) 243–275. [16] J.E. Cermak, Application of modeling techniques to mass transfer and evaporation studies, Paper Presentation, Centennial of Engineering, Chicago, IL, 1952, 15pp. [17] M. Jensen, The model-law for phenomena in natural wind, IngeniNren 2 (4) (1958) 121–128. [18] E.J. Plate, J.E. Cermak, Micrometeorological wind tunnel facility, description and characteristics, Final Report US Army Electronic Research and Development Activity, Fort Huachuca, Arizona (Contract DA-36-0309SC-80371), 1963, 51pp. [19] J.E. Cermak, Wind tunnel design for physical modeling of atmospheric boundary layers, Special Volume on Advances in Fluid Mechanics, J. Eng. Mech., ASCE 107(EM3) (1981) 623–642. [20] A.G. Davenport, N. Isyumov, J.P.C. King, M. Novak, D. Surry, B.J. Vickery, BLWT II: The design and performance of a new boundary layer wind tunnel, in: Proceedings of the Fifth US National Conference on Wind Engineering, Texas Tech University, Lubbock, TX 1985, 2A35—39. [21] T. Miyata, H. Yamada, K. Yokoyama, T. Kanazaki, T. Iijima, M. Tatsumi, Construction of boundary layer wind tunnel for long-span bridges, J. Wind Eng. Ind. Aerodyn. 42 (1992) 885–896. [22] P.A. Irwin, The design of spires for wind simulation, J. Wind Eng. Ind. Aerodyn. 7 (1981) 361–366. [23] N.J. Cook, Simulation techniques for short test-section wind tunnels: roughness, barrier and mixingdevice methods, in: Proceedings of the International Workshop on Wind Tunnel Modeling Criteria and Techniques in Civil Engineering Applications, Gaithersburg, Maryland, Cambridge University Press, London, England, 1982, pp. 126–136. [24] J.E. Cermak, Physical modeling of the atmospheric boundary layer (ABL) in long boundary layer wind tunnels (BLWT), in: Proceedings of the International Workshop on Wind Tunnel Modeling Criteria and Techniques in Civil Engineering Applications, Gaithersburg, Maryland, Cambridge University Press, London, England, 1982, pp. 97–125. [25] W.H. Rae, A. Pope, Low-Speed Wind Tunnel Testing, 2nd Edition, Wiley, New York, NY, 1984, 534pp. [26] J.E. Cermak, Progress in physical modeling for wind engineering, J. Wind Eng. Ind. Aerodyn. 54/55 (1995) 439–455. [27] J.E. Cermak, W.Z. Sadeh, G. Hsi, Fluctuating moments on tall buildings produced by wind loading, Nat. Bur. Stand. Building Sci. Ser. 30 (1970) 45–59.
368
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
[28] T. Tschanz, A.G. Davenport, The base balance technique for the determination of dynamic wind loads, J. Wind Eng. Ind. Aerodyn. 13 (1983) 429–439. [29] D.W. Boggs, Validation of the aerodynamc model method, J. Wind Eng. Ind. Aerodyn. 41–44 (1992) 1011–1022. [30] K. Fuji, K. Hibi, H. Ueda, Anew measuring system using electronically scanned pressure sensors (ESP) and some applications of the ESP system to a square building shape, 9th Symposium on Wind Engineering, Tokyo, 1986, pp. 313–318, (in Japanese). [31] J.A. Peterka, L.S. Cochran, D.W. Boggs, N. Hosoya, M. Downing, Simultaneous peak pressure measurements in the wind tunnel, in: Proceedings of the International Conference on Building Envelope Systems and Technology, Singapore, 1994, pp. 354–359. [32] J. Xie, P.A. Irwin, Application of the force balance technique to a building complex, J. Wind Eng. Ind. Aerodyn. 77&78 (1998) 579–590. [33] A. Steckley, M. Accardo, S.L. Gamble, P.A. Irwin, The use of integrated pressures to determine overall wind-induced response, J. Wind Eng. Ind. Aerodyn. 41–44 (1992) 1023–1034. [34] T.C.E. Ho, G.R. Lythe, N. Isyumov, Structural loads from the integration of simultaneously measured pressures, in: Proceedings of the International Conference on Wind Engineering, Copenhagen, Denmark, 1999. [35] P.A. Irwin, W.W. Kochanski, Measurement of structural wind loads using a high frequency pressure integration method, in: Proceedings of the ASCE Structures Congress XIII, Boston, MA, 1995, pp. 1631–1634. [36] D.W. Boggs, J.A. Peterka, Aerodynamic model tests of tall buildings, J. Eng. Mech. ASCE 115 (1989) 618–635. [37] L.S. Cochran, J.A. Peterka, On breached building envelopes and increased internal pressure, in: Proceedings of the International Conference on Building Envelopes Systems and Technologies, Ottawa, Canada, 2001, pp. 83–87. [38] Y. Uematsu, N. Isyumov, Wind pressures acting on low-rise buildings, J. Wind Eng. Ind. Aerodyn. 82 (1999) 1–25. [39] Y. Tamura, H. Kikuchi, K. Hibi, Wind load combinations and extreme pressure distributions on lowrise buildings, Wind Struct. 3 (4) (2000) 279–289. [40] T. Stathopoulos, Evaluation of wind loads on low buildings — a brief historical review, in: State of the Art Volume, Ninth International Conference on Wind Engineering New Delhi, India, 1995, pp. 155–179. [41] B. Bienkiewicz, Y. Sun, Local wind loading on the roof of a low-rise building, J. Wind Eng. Ind. Aerodyn. 45 (1992) 11–24. [42] Y. Uematsu, N. Isyumov, Peak gust pressures acting on the roof and wall edges of a low-rise building, J. Wind Eng. Ind. Aerodyn. 77&78 (1998) 217–231. [43] J.A. Peterka, N. Hosoya, S. Dodge, L. Cochran, J.E. Cermak, Area-averaged peak pressures in a gable roof vortex region, J. Wind Eng. Ind. Aerodyn. 77&78 (1998) 205–215. [44] F. Beste, J.E. Cermak, Correlation of internal and area-averaged wind pressures on low-rise buildings, in: Proceedings of the Third International Colloquium on Bluff Body Aero. and Applications, Blacksburg, VA, 1996. [45] D. Surry, J.X. Lin, The effect of surroundings and roof corner geometric modifications on roof pressures on low-rise buildings, J. Wind Eng. Ind. Aerodyn. 58 (1995) 113–138. [46] E. Simiu, J.H. Garrett, K. Reed, Development of computer-based model of standards and attendant knowledge-based and procedural systems, in: Proceedings of the Structure and Engineering in Natural Hazard Mitigation, ASCE, New York, 1993, pp. 841–846. [47] A. Rigato, P. Chang, E. Simiu, Database assisted design, standardization, and wind direction effects, J. Struct. Eng. ASCE 127 (8) (2001) 855–860. [48] J. Lin, D. Surry, Simultaneous time series of pressures on the envelope of two large low-rise buildings, BLWT-SS7-1997, Boundary-Layer Wind Tunnel Laboratory, The University of Western Ontario, London, Ontario, Canada, 1997. [49] P.C. Case, N. Isyumov, Wind loads on low buildings with 4: 12 gable roofs in open country and suburban exposures, J. Wind Eng. Ind. Aerodyn. 77&78 (1998) 107–118.
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
369
[50] T. Whalen, E. Simiu, G. Harris, J. Lin, D. Surry, The use of aerodynamic databases for the effective estimation of wind effects in main-force resisting systems: applications to low buildings, J. Wind Eng. Ind. Aerodyn. 77&78 (1998) 685–693. [51] D.W. Boggs, J.A. Peterka, Wind loads on arena roofs using aerodynamic models, in: Proceedings of the Structures Congress ’93, ASCE, Irvine, CA, 1993, pp. 514–519. [52] C.W. Letchford, P.P. Sarkar, Mean and fluctuating wind loads on rough and smooth parabolic domes, J. Wind Eng. Ind. Aerodyn. 88 (2000) 101–117. [53] C. Scruton, On the wind-excited oscillations of stacks, towers and masts, in: Proceedings of the Conference on Wind Effects on Buildings and Structures, National Physcial Laboratory, Teddington, Middlesex, England Symposium No. 16, 1965, pp. 798–836. [54] A. Kareem, T. Kijewski, Y. Tamura, Mitigation of motions of tall buildings with specific examples of recent applications, Wind Struct. 2 (3) (1999) 201–251. [55] L. Singh, N.P. Jones, R.H. Scanlan, O. Lorendeaux, Simultaneous identification of 3-DOF aeroelastic parameters, in: Proceedings of the Ninth International Conference on Wind Engineering, New Delhi, India, 1995, pp. 972–977. [56] N. Isyumov, M. Mikitiuk, Wind tunnel model tests of snow drifting on a two-level flat roof, J. Wind Eng. Ind. Aerodyn. 36 (1990) 893–904. [57] J. Wianecki, A. Chevallier, Excess loads on flat roofs of buildings with and without parapets under the action of wind and snow, in: Proceedings of the Second International Conference on Snow Engineering, Santa Barbara, CA, 1992, pp. 215–228. [58] N. Isyumov, M. Mikitiuk, P. Cookson, Wind tunnel modeling of snow with applications to snow fences, in: Proceedings of the First International Conference on Snow Engineering, Santa Barbara, CA, 1988, pp. 210–226. [59] J.D. Iversen, Small-scale modeling of snow-drift phenomena, in: Proceedings of the International Workshop on Wind-tunnel Modeling Criteria and Techniques in Civil Engineering Applications, Gaithersburg, Maryland, Cambridge University Press, London, England, 1982, pp. 522–545. [60] J.A. Peterka, Wind-tunnel tests — snow deposition and drifting on Denver International Airport Pena Boulevard, Final Report, Cermak Peterka Petersen, Inc., Fort Collins, CO, 1999, 156pp. [61] F.H. Durgin, Pedestrian level wind studies at the Wright Brothers facility, J. Wind Eng. Ind. Aerodyn. 44 (1992) 2253–2264. [62] M.J. Soligo, P.A. Irwin, C.J. Williams, G.D. Schuyler, Acomprehensive assessment of pedestrian comfort including thermal effects, J. Wind Eng. Ind. Aerodyn. 77&88 (1998) 753–766. [63] M.A. Ratcliff, J.A. Peterka, Comparison of pedestrian wind acceptability criteria, J. Wind Eng. Ind. Aerodyn. 36 (1990) 791–800. [64] R.M. Aynsley, W. Melbourne, B.J. Vickery, Architectural Aerodynamics, Applied Science Publishers Ltd., London, England, 1977, 254pp. [65] B.R. White, Analysis and wind-tunnel simulation of pedestrian-level winds in San Francisco, J. Wind Eng. Ind. Aerodyn. 41–44 (1992) 2353–2364. [66] L.S. Cochran, J.A. Peterka, Modeling flow over complex terrain, in: Proceedings of the 2001 Structures Congress and Exposition, ASCE, New York, 2001. [67] J.E. Cermak, Physical modeling of flow and dispersion over complex terrain, J. Boundry Layer Met. 30 (1984) 261–292. [68] R.N. Meroney, Wind-tunnel simulations of flow over hills and complex terrain, J. Ind. Aerodyn. 5 (1980) 297–321. [69] R.P. Hosker, Jr., Flow and diffusion near obstacles, in: Atmospheric Science and Power Production, US Department of Energy, Available as DE84005177 (DOE/TIC-27601) from National Technical Information Service, US Department of Commerce, Springfield, VA 22161, 1984, pp. 241–326. [70] R.L. Petersen, B.C. Cochran, J.J. Carter, Specifying exhaust and intake systems, ASHRAE J. 44 (8) (2002) 30–37. [71] J.B. Wedding, D.J. Lombardi, J.E. Cermak, Awind tunnel study of gaseous pollutants in city street canyons, J. Air Pollut. Control Assoc. 27 (1977) 557–566. [72] J.E. Cermak, K. Takeda, Physical modeling of urban air-pollutant transport, J. Wind Eng. Ind. Aerodyn. 21 (1985) 51–67.
370
J.E. Cermak / J. Wind Eng. Ind. Aerodyn. 91 (2003) 355–370
[73] J.E. Cermak, A.G. Davenport, E.J. Plate, D.X. Viegas (eds.), Proceedings of the NATO Advanced Study Institute on Wind Climate in Cities, Waldbronn, Germany (1993), Kluwer Academic Press, Dordeckt, The Netherlands, 1995, 772pp. [74] P. Kastner-Klein, E. Fedorovich, M.W. Rotach, A wind-tunnel study of organized and turbulent air motions in urban street canyons, J. Wind Eng. Ind. Aerodyn. 89 (2001) 849–861. [75] H.E. Landsberg, The Urban Climate, Academic Press, Inc., New York, NY, 1981, 275pp.
