Early estimation of a structural systems’ stiffness in high-rise buildings A development of guidelines in design for wind Master’s thesis in Structural Engineering and Building Technology HANNA JOSEFSSON DEPARTMENT OF ARCHITECTURE AND CIVIL ENGINEERING CHALMERS UNIVERSITY OF TECHNOLOGY Gothenburg, Sweden 2023 www.chalmers.se www.chalmers.se Master’s thesis 2023 Early estimation of a structural systems’ sti�ness in high-rise buildings A development of guidelines in design for wind HANNA JOSEFSSON Department of Architecture and Civil Engineering Division of Structural engineering Chalmers University of Technology Gothenburg, Sweden 2023 Early estimation of a structural systems’ sti�ness in high-rise buildings A development of guidelines in design for wind HANNA JOSEFSSON © HANNA JOSEFSSON, 2023. Supervisor: Per Langefors, Sweco Examiner: Ignazi Fernandez, Chalmers Master’s Thesis 2023 Department of Architecture and Civil Engineering Division of Structural Engineering Chalmers University of Technology SE-412 96 Gothenburg Telephone +46 31 772 1000 Cover: Illustrative visualization of a high-rise building and the formation of vortex shedding. Typeset in LATEX, template by Kyriaki Antoniadou-Plytaria Printed by Chalmers Reproservice Gothenburg, Sweden 2023 iv Early estimation of a structural systems’ sti�ness in high-rise buildings A development of guidelines in design for wind HANNA JOSEFSSON Department of Architecture and Civil Engineering Chalmers University of Technology Abstract As our buildings grow higher so should the knowledge in the design of high-rise buildings. Tall structures require specialised understanding due to the great impact of wind load and subsequently the importance of stability. In the conceptual phase of design, it is essential to make a good estimation of the required sti�ness. Today’s building standards generally provide a conservative geometry with overestimated loads. A development of guidelines for the design of high-rise buildings in an early stage are desirable and consequently the ambition of this thesis. The aim of the thesis is to examine current methods of design in a comprehensive analysis of high-rise buildings. By evaluating and comparing the design method by building codes with the result from provided wind tunnel testing, new guidelines for early assessment of a high-rise buildings’ structural system will be reached. To accomplish this research di�erent methods of analysis is carried out. To incorpo- rate current design approaches, interviews with structural engineers specialized in tall buildings are conducted. An analytical analysis is conducted by current build- ing codes for design of high-rise buildings followed by a FEM analysis to investigate required sti�ness. The structural system response is evaluated by comfort criteria which set the base for the development of new guidelines. The research showed a distinctly di�erence of wind loads predicted by building standards and wind tunnel testing. This resulted in a discrepancy between required sti�ness for di�erent configurations of the structural system. Observing the results, a development of guidelines for early estimation of a structural systems’ sti�ness in high-rise buildings were reached. Keywords: High-rise buildings, wind, building standard, wind tunnel test, FEM- modelling, sti�ness, early estimation, building design. v Early estimation of a structural systems’ sti�ness in high-rise buildings A development of guidelines in design for wind HANNA JOSEFSSON Department of Architecture and Civil Engineering Chalmers University of Technology Sammanfattning I samma takt som våra byggnader växer sig högre, bör också kunskapen i design av höga hus. Höga konstruktioner kräver specialiserad förståelse på grund av vindens kraftiga inverkan och därmed betydelsen av stabilitet. I en konceptuell fas av de- sig är det viktigt att göra en bra uppskattning av erforderlig styvhet för höga hus. Dagens byggnormer ger generellt en konservativ geometri med överdimensionerade laster. Utveckling av riktlinjer för design av höghus i ett tidigt skede är önskvärt och följaktligen ambitionen med denna forskning. Syftet med avhandlingen är att undersöka aktuella designmetoder i en omfattande analys av höga hus. Genom att utvärdera och jämföra design metoder i byggnormer med resultat från vindtunneltest, kommer nya riktlinjer för tidig uppskattning av höga hus och dess bärverk att utvecklas. För att genomföra denna forskning utförs olika analyser inom höga hus. För att in- tegrera nuvarande metoder av design genomförs intervjuer med byggnadskonstruk- törer specialiserade på höga hus. En analytisk undersökning genomförs av nuvarande byggrekommendationer för projektering av höga hus följt av en FEM-analys för att undersöka erforderlig styvhet. Bärverkets egenskaper utvärderas mot komfortkrav som skapar möjlighet till utveckling av nya riktlinjer. Forskning visade en distinkt skillnad mellan vindlaster som uppskattats av bygg- standarder och vindtunneltest. Detta resulterade i en diskrepans mellan erforderlig styvhet för olika konfigurationer av bärverk. Genom observervation av dessa resultat utvecklades riktlinjer för tidig uppskattning av styvhet i bärverk för höga hus. Keywords: Höghus, vind, byggnadsstandard, vindtunneltest, FEM-modellering, styvhet, tidig uppskattning, byggnadsdesign. vii Acknowledgements This Masters’ thesis focuses on the development of guidelines in design directed towards an early estimation of high-rise buildings. The work of this thesis was performed during spring 2023 as a finalization of the Masters’ program Structural Engineering and Building technology. The thesis is a collaboration between the Building Construction Department at Sweco Sverige AB in Gothenburg and the Department of Architecture and Civil Engineering at Chalmers University of Tech- nology. I would like to show my gratitude to my supervisor at Sweco, Per Langefors for his contributory guidance and advise throughout the work of this thesis. I would also like to address a thanks to my examiner and supervisor from Chalmers, Ignasi Fernandez for his valuable consultation and encouragement for this research. Last but not least I would like to send thanks to the structural engineers Andreas Lin- delöf, Marco Binfaré and Gustav Söderlund for their engagement and contribution to my interview study. Hanna Josefsson, Gothenburg, June 2023 ix List of Acronyms Below is the list of acronyms that have been used throughout this thesis listed in alphabetical order: CFD Computional fluid dynamics DR Drift ratio EC EuroCode EKS EUropeiska konstruktionsstandarder FEM Finite element method HFFB High-frequency force balance HFPI High-frequency pressure integration ISO International Organization for Standardization NBCC National Building Code of Canada SLS Service limit state SS Svensk standard SR Slenderness ratio ULS Ultimate limit state WTT Wind tunnel test xi Nomenclature Below is the nomenclature of parameters and variables that have been used through- out this thesis. Roman upper case letters Aref Reference area B2 Background factor D Along-wind building plan dimension (EI)W T T Sti�ness based on wind tunnel test wind load (EI)EN Sti�ness based on building standard wind load Kx Dimensionless coe�cient L Turbulence length R2 Resonance response factor Re Reynolds number SL non-dimensional power spectral density function T Mean wind velocity time W Across-wind building plan dimension Ẍmax Along-wind acceleration Roman lower case letters b Building width baverage Average building width cdir Directional factor ce Exposure factor cf Force coe�cient cf,0 Fundamental force coe�cient xiii co Orography factor cpe External pressure coe�cient cpi Internal pressure coe�cient cr Roughness factor cseason Season factor cscd Structural factor fL Non-dimensional frequency h Building height hstrip Segment of building height kl Turbulence factor kp Peak factor kr Terrain factor lv Turbulence intensity me Equivalent mass per unit length n1.x Fundamental frequency qb Basic velocity pressure qp Peak velocity pressure r Radius vb Basic wind velocity vb,0 Fundamental basic wind velocity vc Peak wind velocity circular section vm Mean wind velocity vTa Average wind velocity v50 Characteristic fundamental wind velocity w Building minimum width we External wind load z Observed building height zmax Maximum height zmin Minimum height zs Reference height z0 Roughness length z0,II Roughness length for terrain factor II xiv Greek letters ” ”a Logarithmic decrement for aerodynamic damping ”s Logarithmic decrement for structural damping � Transverse deflection ⁄ Slenderness fl Air density ‹ Mean up-crossing frequency ‹k Kinematic viscosity air Âr Reduction factor round corners Â⁄ End-e�ect factor free-end flow ‡ẋ Acceleration standard deviation Ï Solidity ratio „1,x Mode shape xv xvi Contents List of Acronyms xi Nomenclature xiii List of Figures xxi List of Tables xxiii 1 Introduction 1 1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Aim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.2.1 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.3 Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.4 Scope and limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.5 Outline of report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2 High-rise buildings 5 2.1 Why we build tall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2.1.1 Architecture and design . . . . . . . . . . . . . . . . . . . . . 5 2.1.2 Societal consideration . . . . . . . . . . . . . . . . . . . . . . . 6 2.2 High-rise buildings and its structural systems . . . . . . . . . . . . . 7 2.2.1 The core . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 2.2.2 Building loads . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2.3 Design for wind . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2.3.1 Slenderness . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2.3.2 Drift . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.3.3 Building dynamics . . . . . . . . . . . . . . . . . . . . . . . . 11 2.3.3.1 Accelerations . . . . . . . . . . . . . . . . . . . . . . 11 2.3.3.2 Human perception of motion . . . . . . . . . . . . . 12 2.3.3.3 Damping . . . . . . . . . . . . . . . . . . . . . . . . 13 2.3.4 Building aeroynamics . . . . . . . . . . . . . . . . . . . . . . . 13 2.3.4.1 Vortex shedding . . . . . . . . . . . . . . . . . . . . 13 2.4 Building standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.4.1 EN-1991-1-4:2005 . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.5 Wind tunnel test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 2.5.0.1 High-frequency force balance (HFFB) wind tunnel test 17 xvii Contents 2.5.0.2 High-frequency pressure integration (HFPI) wind tun- nel test . . . . . . . . . . . . . . . . . . . . . . . . . 18 2.6 Computational Fluid Dynamics . . . . . . . . . . . . . . . . . . . . . 18 3 Interview study 19 3.1 Interview guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 3.2 Interview I - Andreas Lindelöf . . . . . . . . . . . . . . . . . . . . . . 20 3.2.1 Design of high-rise buildings . . . . . . . . . . . . . . . . . . . 20 3.2.2 A new method for design . . . . . . . . . . . . . . . . . . . . . 22 3.3 Interview II - Marco Binfaré . . . . . . . . . . . . . . . . . . . . . . . 24 3.3.1 Design of high-rise buildings . . . . . . . . . . . . . . . . . . . 24 3.3.2 A new method for design . . . . . . . . . . . . . . . . . . . . . 26 3.4 Interview III - Gustav Söderlund . . . . . . . . . . . . . . . . . . . . 28 3.4.1 Design of high-rise buildings . . . . . . . . . . . . . . . . . . . 28 3.4.2 A new method for design . . . . . . . . . . . . . . . . . . . . . 30 4 Case study 31 4.1 Building A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 4.2 Building B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 4.3 Building C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 4.4 Structural system simplification . . . . . . . . . . . . . . . . . . . . . 40 5 Analytical analysis 43 5.1 Wind load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 5.1.1 External pressure coe�cient for rectangular sections . . . . . . 52 5.1.2 Force coe�cient for rectangular and slender sections . . . . . . 53 5.1.3 Force coe�cient for circular sections . . . . . . . . . . . . . . 54 5.1.4 Development . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 5.2 Acceleration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 5.2.1 Comfort criteria . . . . . . . . . . . . . . . . . . . . . . . . . . 59 6 FEM analysis 61 6.1 RFEM modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 6.2 Building A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 6.3 Building B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 6.4 Building C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 7 Results 69 7.1 Building A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 7.1.1 Wind loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 7.1.2 Frequency and acceleration . . . . . . . . . . . . . . . . . . . . 71 7.1.3 Sti�ness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 7.2 Building B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 7.2.1 Wind loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 7.2.2 Frequency and acceleration . . . . . . . . . . . . . . . . . . . . 74 7.2.3 Sti�ness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 7.3 Building C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 xviii Contents 7.3.1 Wind loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 7.3.2 Frequency and acceleration . . . . . . . . . . . . . . . . . . . . 77 7.3.3 Sti�ness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 7.4 Guidelines for design of high-rise buildings . . . . . . . . . . . . . . . 79 8 Discussion 83 8.1 Interview study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 8.2 Analytical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 8.3 FEM analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 8.4 Development of guidelines for design . . . . . . . . . . . . . . . . . . 85 9 Final remarks 87 9.1 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 9.2 Further investigations . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 References 89 A Appendix Building A I A.1 Wind loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I A.2 Acceleration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V A.3 Sti�ness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI B Appendix Building B VII B.1 Wind loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VIII B.2 Acceleration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XII B.3 Sti�ness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XIII C Appendix Building C XV C.1 Wind loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XV C.2 Acceleration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XIX C.3 Sti�ness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XX xix Contents xx List of Figures 2.1 Slenderness dimensions of a building. . . . . . . . . . . . . . . . . . . 10 2.2 Wind direction illustration. . . . . . . . . . . . . . . . . . . . . . . . . 12 2.3 Illustrative explanation of vortex formation. . . . . . . . . . . . . . . 14 2.4 The Alan G. Davenport chain explanation of wind loading. . . . . . . 16 2.5 Example of a wind tunnel test model. . . . . . . . . . . . . . . . . . . 17 3.1 Final design of Karlatornet at Lindholmen, Gothenburg. . . . . . . . . 20 3.2 Norra tornen in Vasastaden, Stockholm. . . . . . . . . . . . . . . . . 24 3.3 Draken Live at Järntorget, Gothenburg. . . . . . . . . . . . . . . . . . 28 4.1 Entrance floor plan of Building A. . . . . . . . . . . . . . . . . . . . . 32 4.2 Rooftop floor plan of Building A. . . . . . . . . . . . . . . . . . . . . 32 4.3 Section A-A respectively section B-B of Building A. . . . . . . . . . . 33 4.4 Entrance floor plan of Building B. . . . . . . . . . . . . . . . . . . . . 35 4.5 Typical apartment floor plan of Building B. . . . . . . . . . . . . . . . 35 4.6 Section A-A of Building B. . . . . . . . . . . . . . . . . . . . . . . . . 36 4.7 Typical apartment floor plan of Building C. . . . . . . . . . . . . . . . 38 4.8 Roof top floor plan of Building C. . . . . . . . . . . . . . . . . . . . . 38 4.9 Section A-A of Building C. . . . . . . . . . . . . . . . . . . . . . . . . 39 4.10 Structural system of Building A with highlighted core walls. . . . . . . 40 4.11 Structural system of Building B with highlighted core walls. . . . . . . 40 4.12 Structural system of Building C with highlighted core walls. . . . . . . 41 5.1 Map of wind loads zones for fundamental basic wind velocity vb0. . . . 44 5.2 Terrain categories and terrain parameters z0 and zmin. . . . . . . . . 45 5.3 Illustrations of terrain category 0, I, II. . . . . . . . . . . . . . . . . . 46 5.4 Illustrations of terrain category III and IV. . . . . . . . . . . . . . . . 46 5.5 Exposure factor ce(z) for orography factor co(z)=1.0 and turbulence factor kl=1.0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 5.6 Illustrative explanation of reference height zs as per Eurocode. . . . . 50 5.7 Velocity pressure profile for each case of aspect ratio h/b. . . . . . . . 51 5.8 Zones of external wind pressure on a rectangular building. . . . . . . . 52 5.9 Recommended value for external pressure coe�cients . . . . . . . . . 52 5.10 Force coe�cient cf,0 for a rectangular section with sharp corners and no free-end flow. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 5.11 Reduction factor Âr for a section with rounded corners. . . . . . . . . 54 xxi List of Figures 5.12 End-e�ect factor Â⁄ in relation to the solidity ratio Ï and the e�ective slenderness ⁄. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 5.13 Force coe�cient cf,0 for a circular section with no free-end flow. . . . 55 5.14 Development of relation between exposure factor ce(z) and height z for orography factor co(z)=1.0 and turbulence factor kl=1.0. . . . . . 56 5.15 Occupant comfort criteria for vibration acceptability in a building for one-year return period. . . . . . . . . . . . . . . . . . . . . . . . . . . 59 6.1 FEM model of the structural system Building A. . . . . . . . . . . . . 62 6.2 Load models with self-weight and additional vertical loading, respectively. 63 6.3 Dynamic analysis model Building A. . . . . . . . . . . . . . . . . . . 64 6.4 Dynamic analysis model Building A. . . . . . . . . . . . . . . . . . . 64 6.5 FEM model of the structural system Building B. . . . . . . . . . . . . 65 6.6 Load models with self-weight and additional vertical loading, respectively. 66 6.7 FEM model of Building B after dynamic analysis. . . . . . . . . . . . 66 6.8 FEM model of the structural system Building C. . . . . . . . . . . . . 67 6.9 Load models with self-weight and additional vertical loading, respectively. 68 6.10 FEM model of Building C after dynamic analysis. . . . . . . . . . . . 68 7.1 Structural system of Building A with dominant wind direction. . . . . 70 7.2 Evaluation curve for comfort demands. . . . . . . . . . . . . . . . . . 72 7.3 Floor plan of Building B with dominating wind direction. . . . . . . . 73 7.4 Evaluation curve for comfort demands. . . . . . . . . . . . . . . . . . 75 7.5 Floor plan of Building C with dominating wind direction. . . . . . . . 76 7.6 Evaluation curve for comfort demands. . . . . . . . . . . . . . . . . . 78 7.7 Sti�ness reduction factor in relation to building height. . . . . . . . . 80 7.8 Sti�ness reduction factor in relation to building height. . . . . . . . . 81 xxii List of Tables 12table.caption.13 4.1 Structural properties for Building A. . . . . . . . . . . . . . . . . . . . 31 4.2 Wind load properties for Building A. . . . . . . . . . . . . . . . . . . 31 4.3 Frequency for mode 1 to 3 for Building A. . . . . . . . . . . . . . . . 33 4.4 Structural properties Building B. . . . . . . . . . . . . . . . . . . . . . 34 4.5 Wind load properties Building B. . . . . . . . . . . . . . . . . . . . . 34 4.6 Frequency for mode 1 to 3 for Building B. . . . . . . . . . . . . . . . 36 4.7 Structural properties Building C. . . . . . . . . . . . . . . . . . . . . . 37 4.8 Wind load properties Building C. . . . . . . . . . . . . . . . . . . . . 37 4.9 Frequency for mode 1 to 3 for Building C. . . . . . . . . . . . . . . . 39 6.1 Iteration configurations with corresponding loading. . . . . . . . . . . 63 6.2 Iteration configurations with corresponding loading. . . . . . . . . . . 65 6.3 Iteration configurations with corresponding loading. . . . . . . . . . . 67 7.1 Peak wind velocity pressure at the top of the building. . . . . . . . . . 70 7.2 Total wind load and comparison to wind tunnel test. . . . . . . . . . . 71 7.3 Natural frequency from FEM analysis and wind tunnel test. . . . . . . 71 7.4 Frequency and acceleration for each configuration of Building A. . . . 71 7.5 Peak wind velocity pressure at the top of the building. . . . . . . . . . 74 7.6 Total wind load and comparison to wind tunnel test. . . . . . . . . . . 74 7.7 Natural frequency from FEM analysis and wind tunnel test. . . . . . . 74 7.8 Frequency and acceleration for each configuration. . . . . . . . . . . . 74 7.9 Peak wind velocity pressure at the top of the building. . . . . . . . . . 77 7.10 Result and comparison between wind tunnel test and analytical methods. 77 7.11 Natural frequency from FEM analysis and wind tunnel test. . . . . . . 77 7.12 Natural frequency and acceleration for each configuration. . . . . . . . 78 7.13 Sti�ness reduction factor. . . . . . . . . . . . . . . . . . . . . . . . . 79 7.14 Sti�ness reduction factor prediction. . . . . . . . . . . . . . . . . . . . 81 A.1 Properties of Building A. . . . . . . . . . . . . . . . . . . . . . . . . . I A.2 Wind properties at height z, Building A. . . . . . . . . . . . . . . . . II A.3 Wind properties at reference height zs, BuildingA. . . . . . . . . . . . II A.4 Wind loads for segmented division per EN-1991-1-4 method 1A, Build- ing A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II A.5 Wind loads for segmented division per EN-1991-1-4 method 2A, Build- ing A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II xxiii Hanna Josefsson List of Tables A.6 Wind loads on each floor per EN-1991-1-4 method 1B and wind tunnel test, Building A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III A.7 Wind loads on each floor per EN-1991-1-4 method 2B and wind tunnel test, Building A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV A.8 Acceleration data for wind load based on EN-1991-1-4, Building A. . . V A.9 Acceleration data for wind load based on wind tunnel test, Building A. VI A.10 Sti�ness properties Building A. . . . . . . . . . . . . . . . . . . . . . VI B.1 Properties of Building B. . . . . . . . . . . . . . . . . . . . . . . . . . VIII B.2 Wind properties at height z, Building B. . . . . . . . . . . . . . . . . IX B.3 Wind properties at reference height zs, BuildingB. . . . . . . . . . . . IX B.4 Wind loads for segmented division per EN-1991-1-4 method 1A, Build- ing B. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IX B.5 Wind loads for segmented division per EN-1991-1-4 method 2A, Build- ing B. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IX B.6 Wind loads on each floor per EN-1991-1-4 method 1B, Building B. . . X B.7 Wind loads on each floor per EN-1991-1-4 method 2B, Building B. . . XI B.8 Acceleration data for wind load based on EN-1991-1-4, Building B. . . XII B.9 Acceleration data for wind load based on wind tunnel test, Building B. XIII B.10 Sti�ness properties Building B. . . . . . . . . . . . . . . . . . . . . . XIII C.1 Properties of Building C. . . . . . . . . . . . . . . . . . . . . . . . . . XV C.2 Wind properties at height z, Building C. . . . . . . . . . . . . . . . . XVI C.3 Wind properties at reference height zs, BuildingC. . . . . . . . . . . . XVI C.4 Wind loads on each floor per EN-1991-1-4 method 1A and wind tunnel test, Building C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XVI C.5 Wind loads on each floor per EN-1991-1-4 method 2A and wind tunnel test, Building C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XVI C.6 Wind loads on each floor per EN-1991-1-4 method 1B and wind tunnel test, Building C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XVII C.7 Wind loads on each floor per EN-1991-1-4 method 2B and wind tunnel test, Building C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XVIII C.8 Acceleration data for wind load based on EN-1991-1-4, Building B. . . XIX C.9 Acceleration data for wind load based on wind tunnel test, Building C. XX C.10 Sti�ness properties Building A. . . . . . . . . . . . . . . . . . . . . . XX xxiv 1 Introduction This Masters’ thesis is a research of high-rise buildings and their structural systems with the ambition to advance current guidelines of design. This chapter cover an background description along with the aim and scope of this thesis. 1.1 Background Throughout history, high-rise buildings became a symbol for modernization and prosperity. Along with new building technology and the costly and limited space in the cities, high-rise buildings started to qualify as an economic and functional solu- tion. In the beginning of the 20th century, a building of 16 floors was appointed the most profitable height, weighing income against investment cost (Hellemar, 2017). Today, the definition of a high-rise building is distinctly higher. Growing up to over 100 floors, the structure is seen more as a technical di�culty rather than an economic advantage (Caldenby, 1990). Due to its height, the design of high-rise buildings is very complex. Current build- ing codes and guidelines on wind loads are oriented to low-rise buildings and sub- sequently often result in conservative and uncertain geometries of the structural system in high-rise buildings. Further on, building standards generally do not cover important wind e�ects, for instance aerodynamics and crosswind excitation. A reli- able solution to obtain a more realistic load condition due to wind action is to use wind tunnel tests (Irwin, Denoon, & Scott, 2013). However, wind tunnel test are costly and time consuming and consequently only performed in the end phase of the design. A wind tunnel test is generally used in a detailed end design to reduce overestimated material as well as being a confirmation of the earlier design choices. To make an assessment of a high-rise buildings’ structural system in early stage, FEM modelling is commonly utilized. This method brings the results substantially closer to the wind tunnel test outcome in comparison to static calculations by build- ing standards. FEM modelling of high-rise buildings can be performed in various modelling programs and in diverse level of detail. The development of structural assessment and building design have over time been the result of pushing the limits and striving towards the impossible (Isaksson, Mårtensson, & Thelandersson, 2020). In early planning of high-rise buildings, it is crucial to make a good estimation of how much space the stabilizing structural 1 1. Introduction system will take in order to design the floor plan. To facilitate the design process of high-rise buildings, particularly in early stages, an improvement of current building standards is desirable and is the ambition and motivation for this thesis. 1.2 Aim This Masters’ thesis aims to be incorporated in new guidelines to use in early esti- mations of high-rise buildings in order to make better approximations of how much space the structural system will take. By critically identify and evaluate di�erences in current design techniques, new understanding that is crucial for the improvement of the design process of high-rise buildings will be reached. 1.2.1 Objectives The aim of the thesis will be achieved by the accomplishment of the following ob- jectives: • Identification of di�erent methods for design of high-rise buildings by perform- ing a literature study. • Interviews with structural engineers specialized in the design of tall buildings. • Implementation of principal calculations for high-rise buildings by building standards including comparison with wind tunnel test result. • FEM analysis of three reference buildings, conducting an eigenfrequency in- vestigation to use in comfort criteria evaluation. • Development of methods on early estimation of a structural systems’ sti�ness based on an assessment of previous conclusions. 1.3 Method The initial phase of the thesis consists of a comprehensive literature study on the design methods of structural systems for high-rise buildings. The literature study is performed to give the thesis a foundation and to be able to integrate obtained knowledge in the proceeding development of the work. To gather, document and possibly incorporate current design approaches within the industry in this research work, interviews with high-rise building designers are car- ried out. The goal with these interviews is to assimilate di�erent views of the current techniques of designing high-rise buildings, identify what is missing and highlight what is in need of development. 2 1. Introduction Three di�erent high-rise buildings will be used as case study for the development of the thesis. Following, a comparison is made for each building where the wind loads is calculated analytically by building standards respectively extracted from provided wind tunnel testing. This will allow to identify the gap between the di�erent design approaches for tall buildings. To analyze the buildings’ response and subsequently extract required sti�ness of the structural system, finite element programs are used. By critically evaluating and comparing existing knowledge together with results gathered from the FEM analysis, the desired development of methods is to be es- tablished. 1.4 Scope and limitations This thesis is an investigation of high-rise buildings. The thesis is provided with three di�erent reference buildings that are observed and utilized for developing new guidelines for deign. The buildings observed are each of di�erent height and con- figuration. This allow for a comprehensive study of high-rise buildings of various designs. The analysis will be limited to these reference buildings and the final con- clusions are based on these results. Wind tunnel tests and architectural drawings relative to each reference building are provided by Sweco Sverige AB. Di�erent buildings code standards appurtenant to di�erent regulations regarding wind loads and comfort demands. Although a diversity of methods would present a broader perspective, this thesis is limited to Eurocode European standards (EN- 1991-1-4, 2005) together with additional standards by International Organization for Standardization (ISO:10137, 2008) and Building construction recommendations EKS:12 (BFS:2022:4, 2022). Acceleration and frequency calculations are only per- formed for along-wind accelerations, as described in EN-1991-1-4. The Eurocode does not consider across-wind acceleration and neither does this thesis. FEM modeling are conducted in the finite element program RFEM and the models are designed simplified where the main structural system alone is assumed to carry all the load. The slabs are modelled with an simplified geometry as the loading are manually concentrated on the core walls. The core walls are modelled with fewer openings than in reality which needs to be considered in the end results. Furthermore, the material of the structures is assumed to be entirely made out of concrete and behave linear elastic. Although di�erent concrete qualities are used in each building, the structural systems are modelled using an average concrete class of C40/50. The sti�ness is collected by a section generator in the program FEM- Design where the simplified cross sections are sketched. In design of high-rise buildings the influence of piles are crucial, especially for struc- tures mounted at ground conditions with geotechnical di�culties. Wind loads act- ing on the building are transferred down to the piling foundation. Subsequently, the sti�ness of the piles are of great importance in design. However, the design of piles are of a completely di�erent investigation and hence not discussed in this thesis. 3 1. Introduction Tall structures are designed based on the impact of wind and seismic activity. The buildings considered in this thesis are based in Sweden where seismic activity is not an issue. Of this reason, seismic consequences are not considered. If development of guidelines for the design of high-rise buildings are to be conducted in a location where seismic activity is crucial, this needs to be addressed. 1.5 Outline of report The thesis is divided into several chapters which added together provides for the development of guidelines for design. Chapter 2 Literature study of high-rise buildings and how to design for wind. Chapter 3 Interview study with designers specialized in high-rise building design. Chapter 4 A case study of three di�erent high-rise buildings. Chapter 5 The theory of building standards are presented and explained. Chapter 6 Description of FEM analysis conducted for each case building. Chapter 7 Presentation of results from analytical analysis and FEM modelling. Chapter 8 Discussion of results. Chapter 9 Final remarks with conclusion and proposal for further investigations. 4 2 High-rise buildings 2.1 Why we build tall As the world grow, so does our buildings. Throughout history, there have always been an interest in building towards the sky. Building tall advocate for wealth and power and have been a tool for promoting prosperity. In present time, we build tall to make money. High-rise buildings that generate further floor area while uti- lizing the simple, repeating building design that is a natural choice for skyscrapers would enlarge revenue. However, the costly space together with construction and maintenance expenses needs to be compensated by the rent income. Subsequently, the buildings must be suitable for people meant to use them. The most profitable design to make money would be to maximize the using space inside whilst keeping the construction and architecture cost at a minimum (Ascher, 2011). In decision of building height, there is an economic height considered which give the highest return of investment cost. A buildings’ e�cient height is strictly depending on the design of the structural system. Even if a taller building contribute to extra floor space, higher buildings also require a stronger core, additional bracing and larger mechanical systems that all occupy an extensive part of the buildings’ space area. Since the commercial viability is as vital as the structural design itself, the buildings’ height are rarely adapted to its e�cient height (Ascher, 2011). From an economic perspective, a minimum net floor area of 75% is needed for a building to be regarded as e�cient in its use (Sarkisian, 2016). The definition of a high-rise building is diverse. One interpretation is that a building can be labeled a high-rise building when the height exceeds 23 meters (Sarkisian, 2016). Another characterization is to regard a building as tall when the aspect ratio exceeds 5:1. All together it can be stated that a building is considered a high-rise when its height a�ect the design substantially (Truby, 2014). 2.1.1 Architecture and design Building aesthetics are often a secondary consideration due to the importance of other aspects such as construction and safety. However, the design of high-rise buildings is an iterative process between structural engineering and architecture (Ascher, 2011). Skyscrapers has a vital impact on its surroundings and its design needs to match its location and purpose. Due to economical reasons, the structure of high-rise buildings often becomes the architecture itself (Ascher, 2011). 5 2. High-rise buildings The initial phase of designing a high-rise building is to select a core. The location of the core can vary and together with its composition it will have a huge impact on the floor layout. The core act as the main structural component which provide the building with required sti�ness to resist lateral forces (Ascher, 2011). A valuable part of high-rise buildings is simplicity. By designing for purity, the structure will response in harmony with the architecture (Sarkisian, 2016). Often in high-rise buildings, the structural design can be translated by looking at the architecture. By designing the structure to follow the natural flow of force, the building becomes the most e�cient in its use. The force always takes the shortest and easiest way through the structure (Sarkisian, 2016). 2.1.2 Societal consideration Super-tall buildings today are of multi-use; residential, o�ce space, hotels, gyms etc., making each skyscraper an individual "vertical city". In toady’s society, people spend approximately 90% of their day inside, independently on activity. Conse- quently, the buildings we spend our lives in a�ect our health (Ascher, 2011). The wind load acting in the building, do not only bring technical issues of deign, but the sway induced by wind becomes a problem for the users of the building. The movements generated by the wind could potentially cause motion sickness and dis- comfort for the users. In the US, the limit for sway is 1/500th of the building height (Ascher, 2011). While in Europe, there is only recommendations for maximum top deflection limit. The floor plan of the building needs to be design in line with its purpose. For high-rise buildings the usage is typically of either residential, hotel, commercial or retail use. Although every individual type of use needs to be addressed through design with di�erences in height, acoustic measures, and spacing of walls, the main dominating discrepancy is the occupancy rate. For residential use of a building with an area of 1000m2 the occupancy rate would land between 20-30 people whilst hotel use would be 35-40 people and 80-120 people for o�ce use (Truby, 2014). Since the human is more sensitive to motion when laying down, the limit of ac- celeration is lower for buildings of residential use rather than o�ce or commercial based spaces. For one year horizontal acceleration return wind period, the criteria for an o�ce space is 10-13 milli-g’s compared to a residential of 5-7 milli-g’s. These limits are based on a damping ratio of 1,5% for concrete and a maximum predicted acceleration from wind impact (Sarkisian, 2016). 6 2. High-rise buildings 2.2 High-rise buildings and its structural systems When designing a high-rise building, the same design approach as for low-rise build- ings is made, but with some additional considerations. Specialists in di�erent pro- fessions frequently needs to be involved in the process such as wind, geotechnical and fire specialists. A significant di�erence from designing low-rise buildings is the impact of wind and the lateral loading due to this natural phenomenon causes. Since the wind load is very dominant for high-rise buildings, this load case usually governs the design and choice of structural system, especially the selection of main stabiliz- ing units such as core, columns and walls (Truby, 2014). In this regard sway of the building becomes an important requirement di�cult to meet. To prevent excessive swaying of the building it is fundamental to provide the structure with a bracing system. The lateral bracing of the building gives the struc- ture its required sti�ness and restrict movements induced by wind loads and other lateral loads (Ascher, 2011). Furthermore, the structural framing system will not only provide the building with its needed strength and sti�ness but also determine the structures’ responses to drift and accelerations (Truby, 2014). Besides the structural system itself, dampers are commonly used to shift weight around in order to prevent heavy sways caused by wind (Ascher, 2011). To obtain a high degree of e�ciency of the structures’ sti�ness, the structural elements should be placed at the buildings perimeter. To acquire the most e�cient sti�ness in a high-rise building, a squared shaped structure with large columns in each corner generates the best result in this manner. While for a circular shaped structure with columns at its edges the e�cient sti�ness is reduced to 50% in comparison due to increased dynamic behaviour (Sarkisian, 2016). 2.2.1 The core The core is the fundamental part of a buildings’ structural system. It will accom- modate all vertical movement such as elevators and stairs. Most importantly it will be the dominant component resisting lateral loading (Truby, 2014). Due to its high importance, the core needs to be designed in a precise and optimized way. An ine�ective design will increase the final cost since it will a�ect both the usage of floor area and the quality of the structural system, and therefore aggravate the construction process (Truby, 2014). Using core shear walls for stability, e�ciency is found when locating these walls symmetrically in relation to the centre lines of the structure. Increased lengths of the core walls will also provide e�ciency to the structure. Often, the thickness of the core wall is situated between 350mm to 600mm when observing buildings up to 200m in height. Due to this considerable thickness, it is crucial to make detailed approximations in initial phase of design in order to develop the floor plan in collaboration with architects (Truby, 2014). 7 2. High-rise buildings 2.2.2 Building loads In early design it is vital to design for service limit state, SLS, which is often the gov- erning design aspect. However, ultimate limit state, ULS, also needs to be analysed. In high-rise buildings, wind loads are governing for the design of high-rise buildings (Truby, 2014). Seismic loads is not considered in Sweden due to its non-critical seismic ground. The structural system is designed based on the loads acting on the building. There are sustained loads (permanent load) from the buildings self weight and transient load (variable load) from forces acting on the building. The permanent load is of maximum certainly whilst the transient loads depends on the behaviour of the sys- tem. When designing for high-rise buildings, load combinations of transient and sustained load is accommodated by reduction factors, weighing up for the probabil- ity that the worst loads will not occur simultaneously (Sarkisian, 2016). In careful design, the gravity loads from the structure can be favorable and used to resist overturning moments caused by imposed loading (Sarkisian, 2016). In-situ reinforced concrete is often used when building tall. Since concrete have a large amount of self-weight, the structure alone helps with the stability (Truby, 2014). A rough estimation is that as the height of the building is doubled, so is the total gravity load (Ascher, 2011). The material have also natural damping that resist sway. The natural damping of a material is defined in how well the material is on disperse energy (Truby, 2014). The dominant load a�ecting the design of high buildings are the lateral load caused by wind (Ascher, 2011). The load from wind acting on the building facade, increases with the height of the building, exponentially (Sarkisian, 2016). When doubling the height of the building, the lateral load caused by wind forced are increased with a factor of four (Ascher, 2011). Wind will give rise to a dynamic e�ect of the building since the load is not static. The load is added in di�erent degrees and unpredictable times, even though one can predict an roughly repeating cycle of wind motions (Truby, 2014). The e�ect of wind loads could be damaging of the structure. It can also create discomfort to the buildings’ users. The building can also cause unpleasant wind motions at street level. 2.3 Design for wind Wind is a complicated phenomenon that acts in three dimensions and needs to be considered over time. When designing for wind, the wind load prediction is mainly gained from observational studies of the current wind climate on site. The phenom- ena wind is both depending on the orography and temperature di�erences at the surface (Hughes, 2014). 8 2. High-rise buildings Wind load is a variable load that is mainly a�ecting the external surfaces of a building, but also create an internal pressure. Measured in force per surface area, the wind load is assumed to be directed perpendicular to the buildings’ facade. Generally, structures today possess enough mass and natural damping to resist mo- mentary loading of the dynamic wind force (Isaksson et al., 2020). The wind load pressure and correspondingly the wind velocity is increased gradually with height of the building facade. Above some point, depending o the building design ad location, the wind speed is assumed constant (Hughes, 2014). The orientation and shape of the structure is of great importance when designing a high-rise building for wind movement (Sarkisian, 2016). Wind pressure acting on a building depends on di�erent aspects, wind direction and speed as well as the shape of the building. Wind is seldom uniform, the speed increased at higher height but is more predictable. To reduce the lateral pressure on the structure, the building can be designed with holes in it (Ascher, 2011). Wind acting directly on the windward side of the building has the greatest impact on the structure. The negative suction pressure that occurs on the leeward side and adds on to the loading together with drag e�ects parallel to the wind direction. The wind direction, magnitude and the topography is di�erent for all sites and con- tribute to change in designing for wind. Fundamental observation in design in the strength of the structure along with its serviceability. High-rise buildings are most commonly designed looking at across-wind rather than along-wind. This is due to the phenomena vortex shedding that generate normal forces to the perpendicular side of the buildings (Sarkisian, 2016). The wind impact on the building is strongly influenced by the topography of the location. The resulting wind pressure is substantial higher in an open area in com- parison to a building dense area (Sarkisian, 2016). 2.3.1 Slenderness When designing the building for sway the slenderness of the structure is of high importance. Especially at the initial phase of design, the slenderness ratio can in- terpret how active the structural system will be (Truby, 2014). The aspect ratio, or slenderness ratio (SR) defines the relation between height and minimum width of the structure, see Equation 2.1 and is desired to fall between 6/1 or 7/1. Yet it is not uncommon for high-rise buildings to obtain value of 8/1. In such cases, damping is of great importance for comfort reasons (Ascher, 2011). When the slen- derness ratio exceeds 8/1 the dynamic performance of the structure will dominate the design. However, the slenderness ratio should only be used in the beginning of design for approximation. For final design, widths of the structural system will be the governing dimensions (Truby, 2014). Slenderness dimensions are illustrated in Figure 2.1. 9 2. High-rise buildings SR = h/w (2.1) where h building height [m] w building minimum width [m] Figure 2.1: Slenderness dimensions of a building. 2.3.2 Drift Wind acting on the building will give rise to a lateral displacement of the building called drift. The drift is generally defined as the di�erence in displacement at the very top of the structure but can also be measured locally between floors (Sarkisian, 2016). If the lateral displacement is too high for each floor, this could lead to poor consequences for the design of the facade where cladding and internal partitions could be a�ected (Truby, 2014). The limitation for drift ratio (DR) in design are often set to 1/500 of the structure height when analyzing the elastic response and is calculated by Equation 2.2. Even though, some buildings with a drift of 1/400 has been constructed (Sarkisian, 2016). The inter-storey drift is approximately between 1/500 and 1/200 (Truby, 2014). DR = �/h (2.2) where � transverse deflection [m] h building height [m] 10 2. High-rise buildings When the building is objected to wind that generates lateral forces, second order e�ects needs to be considered. The consequence of second order e�ects could raise the drift with as high as 10% (Sarkisian, 2016). 2.3.3 Building dynamics When building tall, considerations regarding the buildings dynamic performance needs to be incorporated. Wind loads of di�erent frequencies will a�ect the build- ing in diverse ways. Along with the initial damping in construction materials, the structure will be impacted by its natural frequency. If the frequency generated by wind is approaching the natural frequency of the structure, there is a possibility of consequence regarding heavy loading and displacement (Truby, 2014). Tall buildings in general conduct low natural frequencies which makes the structure more susceptible to be influenced by wind (Kwok, Hitchcock, & Burton, 2009). An approximation of a buildings fundamental frequency is generated by taking the number of floors divided by 10. To analyze the fundamental frequency period more closely, each mode needs to be observed (Sarkisian, 2016). 2.3.3.1 Accelerations Heavy winds generates movements of the structure. Acceleration is generated by both along- and across-wind motions. For very slender buildings such as skyscrap- ers, across-wind accelerations tends to dominate the choice of design. The biggest impact a�ecting the acceleration is the shape of the building along with its height and location (Sarkisian, 2016). To confirm the dominance of across-wind acceleration over along-wind, Equation 2.3 needs to be fulfilled. The di�erent wind directions are illustrated in Figure 2.2. Ô W · D/h < 1/3 (2.3) where W cross-wind building plan dimension [m] D along-wind building plan dimension [m] h along-wind building plan dimension [m] 11 2. High-rise buildings Figure 2.2: Wind direction illustration. 2.3.3.2 Human perception of motion Maximum acceleration due to resonant response generated by wind load occur at the top of the building and can cause an unpleasant perception of motion at the uppermost floors. As humans, we can easy feel a discomfort when experiencing vibration responses in a building. However, the sensitivity to vibrations are di�erent for everyone and depends on several aspects such as the height of the building, acoustics and visuals. From investigations regarding the perception of motions, majority of building users would feel the dynamic response of a building subjected to an acceleration greater than 0.1 m/s2. Table presents an idea of human perception to di�erent magnitudes of acceleration. (Abu-Zidan, Mendis, Gunawardena, Mohotti, & Fernando, 2022). Table 2.1: Human perception to building acceleration (Abu-Zidan et al., 2022). Level Acceleration [m/s2] Human perception 1 < 0.05 No perception of motion 2 0.05 - 0.1 Sensitive users perceive motion 3 0.1 - 0.25 All users generally perceive motion 4 0.25 - 0.4 O�ce work impossible 5 0.4 - 0.5 Di�culties when walking naturally 6 0.5 - 0.6 Walking naturally is impracticable 7 0.6 - 0.7 Impossible to walk 8 > 0.85 Falling objects in the building After repeatedly exposure to high vibrations, users of the building could experience several issues such as nausea, dizziness and headache (Kwok et al., 2009). Human sense of motion is investigated by looking at the buildings’ behavior, sti�ness and damping (Sarkisian, 2016). Since the users comfort of the building is a crucial consideration, accelerations and top deflections of a high-rise building are generally dominating the design choices (Abu-Zidan et al., 2022). 12 2. High-rise buildings 2.3.3.3 Damping To regulate the acceleration movement of the building, damping is used. Damping of the structure is due to the material used and in relation to the return period for wind load. For concrete the damping ratio of a 50 year return period is 3% (Sarkisian, 2016). Damping reduce the lateral acceleration. This is important for the users perception of building motion. To reduce the lateral acceleration, natural damping from use of material can be adjusted by changing the structure design, shape, adjusting aero- dynamic response (Truby, 2014). Damping correspond to magnitude, hence with stronger winds the displacements will be larger. Which damping value that should be chosen in design is not set, and the opinion di�er from engineer to engineer. Total damping values between 1.5-3.0 % of critical have earlier been used for design (Truby, 2014). states that the upper limit is probably an over-estimation of damping value when looking at ultimate-limit state. 2.3.4 Building aeroynamics Non-aerodynamic shapes as squares give rise to a wind phenomena called Vortex shedding. When wind meets the face of the building, the pressure di�erence at the di�erent sides of the building creates vortices that that pressure the building perpendicular to the wind direction. Can be a problem at high frequencies. To minimize vortex shedding, the building can be placed so that the long facade is located parallel with the wind direction. There is also design measures to prevent this phenomena, di�erent shapes such as rounded corners, twists and holes create a more aerodynamic structure (Ascher, 2011). 2.3.4.1 Vortex shedding As previously mentioned, across-wind is governing for high and slender buildings. This wind motion perpendicular to the wind direction generates a phenomenon called vortex shedding (Sarkisian, 2016). Vortex shedding is a wind motion of flow that generates asymmetric pressure distribution acting on the across-wind facade of the building, illustrated in Figure 2.3. If the frequencies induced from vortex shedding occur spontaneously as the natural frequency, resonance would develop (Abu-Zidan et al., 2022). The e�ect of vortex shedding very much depends on the shape of the building. A circular shaped high-rise building will give rise to a higher cross-wind motion due to vortex shedding than a rectangular one. Further devel- opment that reduces vortex shedding is to integrate holes in the building structure. Implementing the vortex shedding frequency along with the building shape and wind speed one can define a so called strohaul number that characterizes the oscillating flow mechanism (Sarkisian, 2016). 13 2. High-rise buildings Figure 2.3: Illustrative explanation of vortex formation. 2.4 Building standards Building standards translate the complex wind movement into simplified loads that can be more easily understood (Sarkisian, 2016). Building codes are today very conservative since they are based on general wind conditions and with a simplified building shape in consideration.Design approaches of wind based on current stan- dards is established from historic climate data of the specific area (Sarkisian, 2016). The codes do not account for the aerodynamics of the building (Ascher, 2011). Nei- ther do the majority of building standards of today not perpetrate to the common design criteria for drift (Sarkisian, 2016). With respect for di�erences in di�erent codes, the majority of building standards is applicable to low-rise buildings and often only generates wind loads in along-wind direction. This is a very conservative demarcation since other aspects from wind loads needs to be considered in design, such as crosswind and torsional loading (Irwin et al., 2013). In its nature, the motion of wind is dynamic and needs to be considered in the design of buildings. If a building acquire great sti�ness and damping, the structure can be seen as static and subsequently designed in a simplified manner (Boverket, 1997). 2.4.1 EN-1991-1-4:2005 The Eurocode Eurpoean standards includes methods used for verification of a build- ings’ capacity and strength. For guide on determining wind loads in design of build- ings, Eurocode EN 1991-1-4:2005 is used. The Eurocode states to issue methods for design of buildings and structures of both traditional and innovative character, however the standard is inadequate for more unusual conditions and the code refer to adopt specific investigations. The code also refers to further investigations of the design using wind tunnel tests or detailed numerical design methods (EN-1991-1-4, 2005). The Eurocode is applicable to buildings not exceeding 200 m in height and gives guidance for natural wind loads against every observed surface of the building, in- cluding attached components (EN-1991-1-4, 2005). It is fundamental to specify predicted wind loads for each mean recurrence interval along with its critical is- 14 2. High-rise buildings sues. These wind loads along with partial factors accounting for the unpredictable events are achieved using existing building codes. Although building codes may be an adequate way to get a rough estimation of the design of high-rise buildings, the standards lack quality in including wind phenomena as crosswind excitation, aerodynamics and the e�ect of neighbouring buildings. Absence of these may not only lead to an ine�cient design with higher costs but also malpractice of building motions that could a�ect the users of the building (Irwin et al., 2013). The current Eurocode do not incorporate wind maps and is therefore meant to be adapted together with a national annex compatible to the wind climate in question. Presented method and equations in the Eurocode generate characteristic values of wind loads that apply directly to a service limit state analysis. To validate the building in ultimate limit state, additional partial factors would be needed (Hughes, 2014). 2.5 Wind tunnel test Wind tunnel tests have been a part of the building industry since the rise of Word Trade Center in New York in the 1960s and have been developed ever since. To analyse the structural loads generated by wind and the buildings’ response to these, wind tunnel test is carried out. Building code calculations are inadequate to specify the details needed for designing high-rise buildings and consequently wind tunnel testing is needed (Irwin et al., 2013). According to Sarkisian (2016), buildings higher than 40 floors should be checked in a wind tunnel test. According to Irwin et al. (2013), a wind tunnel test would be favorable for a building that satisfy any of following criteria: i. The height of the building h, exceeding 120 meters. ii. The height of the building h, is greater than four times the average width baverage. iii. Minimum natural frequency is smaller than 0.25 Hz. iv. The abbreviated wind velocity v/(f1baverage) at ultimate limit state is larger than five. Subsequently, the need of a wind tunnel test is depending on many aspects. Each building is di�erent in its design with varying slenderness, location and structure, which all will determine the obligation of a wind specialist involvement (Irwin et al., 2013). Although wind tunnel tests are frequently conducted in the design of high-rise build- ings, the engineers implementing the results rarely have enough knowledge to un- derstand it. If correct used, the wind loads gathered from wind specialists and wind 15 2. High-rise buildings tunnel test can minimize the cost of deign significantly. It is vital that the designer using these wind loads generated by wind specialist with a critical eye. Similarly to the possibility of di�erent interpretation of results by the engineers, di�erent wind consultants will generate diverse wind loads. Subsequently is is important as engi- neer to have knowledge in possible cause of uncertainties that could arise in a wind tunnel test (Irwin et al., 2013). To be able to implement the wind loads generated by a wind tunnel test in an accurate manner, five steps should be conducted. This method is described by Alan G. Davenport as the "chain", see Figure 2.4. We have all heard of the expression "the chain is never stronger than its weakest link", and the same apply to a wind tunnel test implementation. The chain can be used both for an analytical method as well as for a wind tunnel test procedure. The most important requirement of using the chain is to remember to consider the two di�erent chains separately. Similarly as the process will be weakened if one link of the chain is neglected, the result will not be accurate if one is to combine an analytical method together with results from a wind tunnel test. The two di�erent approaches is only to be compared and evaluated against each other in the end (Irwin et al., 2013). Figure 2.4: The Alan G. Davenport chain explanation of wind loading. The first part of the chain is the wind climate where the direction and velocity of the wind at the specific location is required. Secondly, the influence of terrain is considered. This part of the chain includes the topography together with the sur- face roughness. Aerodynamic e�ects is the next link which incorporate the local e�ects from neighboring buildings. The dynamic e�ects of the chain speak for the building movement generated by wind, also incorporating aeroelastic consequences. The last link of the chain is represented by the importance of criteria, to estimate the building under the e�ect of wind (Irwin et al., 2013). When choosing a building design, the structure needs to be checked against prevail- ing wind at its intended location. This is made by using wind-tunnel tests where the building design is iterated several times to find and e�cient solution. In the wind-tunnel test, the strength and performance of the building in relation to the wind impact is analyzed (Ascher, 2011). When creating a wind-tunnel test, all buildings within a radius of approximately 0.8 km is constructed in a miniature version, usually on a scale of 1:400. The model is put on a rotational platform, see Figure. The wind is simulated by fans with a speed up to 100 km/hour. To see the vortices formation, smoke is used (Ascher, 2011). 16 2. High-rise buildings Figure 2.5: Example of a wind tunnel test model. According to Sarkisian (2016), a wind-tunnel test should consist of following inves- tigations: I. Wind model simulation of the building in question together with surrounding construction within a radius if 0.8 km based on climate data. II. Exterior walls modelled with pressure taps to provide for pressure measure- ments. III. Wind analysis for pedestrian comfort. IV. Force-balance simulation. V. Aero-elastic simulation for height at minimum 300 m. When creating a wind tunnel test, the specific topography is modelled with neigh- bouring buildings and relevant landscaping a�ecting the wind motion. Wind tunnel testing is of two di�erent types, high frequency .. and cladding pressure design, where the first mentioned is the most common performed, generating wind loads needed to design the stabilizing system of the structure. The inferior one is where local pressure is analysed on di�erent parts of the buildings’ face in order to design the facade (Irwin et al., 2013). Both types of wind tunnel testing is presented in following chapter. 2.5.0.1 High-frequency force balance (HFFB) wind tunnel test In a high-frequency force balance (HFFB) wind tunnel test, a rigid model of the structure is connected to a balance that rotates the building model for specific 17 2. High-rise buildings angles and wind speeds. The wind tunnel test measure reactions in the base that is used to generate corresponding wind loading on each floor as well as accelerations and deflections (Abu-Zidan et al., 2022). 2.5.0.2 High-frequency pressure integration (HFPI) wind tunnel test To optimize the building facade against wind loads, the wind tunnel test high- frequency pressure integration (HFPI) is utilized. The model is covered in an amount of pressure taps, recommended at least one tap per 120 m2 of the facade. This method observe instantaneous pressures at local building areas which are used to design customized facade elements (Abu-Zidan et al., 2022). Subsequently, the cost of design could be minimized considerable. 2.6 Computational Fluid Dynamics Computational fluid dynamics (CFD) is a new developed method for analyze of the building response induced by wind. It is specifically adopted to predict ground-level wind velocity (EN-1991-1-4, 2005). This numerical method simulate fluid flow based on the fundamental equations of Navier-Stokes and is a favorable parametric tool that is cheaper and less time consuming compared to a wind tunnel test. Using CFD, more advanced architectural designs can be evaluated (Abu-Zidan et al., 2022). An advantage of using CFD is that the technique includes the e�ects of temperature which are not conducted in building codes. In comparison to a wind tunnel test, a CFD model measures infinitely amount of fluctuations on the building while a wind tunnel test is narrowed down to the amount of pressure taps mounted on the building model (Thordal, Bennetsen, & Koss, 2019). Unfortunately, this method is not yet competent enough to generate quantitative results of detail for wind loads (EN-1991-1-4, 2005). The reason for this uncertainty is that the CFD model demand the designer to set several parameters which subsequently is a risk due to possible miscalculations. However, CFD is a advantageous method to use in an early conceptual stage of design for high-rise buildings (Abu-Zidan et al., 2022). 18 3 Interview study To incorporate current design approaches in the design of high-rise buildings, a few interviews with structural engineers specialised in tall buildings are conducted. Each interview and subsequently designer is connected to a specific reference building that puts the interview in a narrower context of high-rise buildings. 3.1 Interview guide The interviews are performed following an interview guide accommodating general questions in the design of high-rise buildings. The interview guide contains the fol- lowing aspects: I. How is the design process in early stage for high-rise buildings carried out? II. Is the assessment of design in early stages based on analytical calculations or FEM modelling? III. What is the general time span of design in early stage? IV. How is the workflow between architect and building designer? V. What guidelines or industry practise are implemented in design of high-rise buildings? VI. Are frequency and acceleration calculated to evaluate a structural systems’ sti�ness and stability? Are there other parameters observed in assessment of stability? VII. What wind load models are used in early design of high-rise buildings? VIII. What is absent or in need of development in current codes for early design of high-rise buildings? IX. In development of this thesis and its research, what is relevant to examine and what is desired to gain in a new method of early deign of high-rise buildings? To personalize and modify each interview to what is relevant for each specific de- signer, the questions are only used as a initial guidance rather than a strict scheme. Subsequently, gathered material from each interviews is summarized in relevant sub- heading and not each question from the interview guide. 19 3. Interview study 3.2 Interview I - Andreas Lindelöf The first interview is with Andreas Lindelöf who is a strcutural engineer at VBK Konsulterande ingenjörer AB in Gothenburg. Reference building used in this in- terview is Karlatornet at Lindholmen, Gothenburg, see Figure 3.1. After its final construction, Karlatornet will be a 246 meter high skyscraper located next to the Gothenburg harbor (serneke.se, 230515). The interview with Andreas was performed at VBK’s o�ce in Gothenburg. Figure 3.1: Final design of Karlatornet at Lindholmen, Gothenburg. 3.2.1 Design of high-rise buildings The design of high-rise buildings at VBK is a parameterized based work where a FEM model is created in an early stage. The FEM program used is ETABS with appurtenant script in Excel and Visual Basic. High-rise buildings are often repeti- tive in its design and hence easy to model. Andreas describes that the early models usually takes one or two days to create, depending on the amount of set-ups along with its mesh size. The model is a pretty rough estimation of the structural system with some simplifications, but still more detailed than general FEM models in the building industry. When analysing the model, di�erent thicknesses and geometries of the core are observed. Several parameters are measured in these iterations; fre- quency, acceleration, total horizontal displacement at the top of the building and concrete weight. The results are visualised in tables and diagrams that makes it easy to quickly appoint a favorable solution. 20 3. Interview study The design of high-rise buildings is an iterative process together with the architects. Except for a design including outriggers, the columns in the buildings’ structural system has no impact on the stability and are only positioned in the FEM model as a representation in early design. The floor layouts with their structural systems are chosen together with the architects. Several FEM models are analysed before taking an informative decision based on the resulting data set. Based on experience, the wind loads considered in design are set pretty low, varying depending on project. Andreas describes that the wind loads usually are adjusted to terrain category IV based on the Eurocode European standards and that they tend to be even more o�ensive regarding the exposing factors. This method is not based on a strict scientific rule, but implemented to reduce the wind loads. The natural frequency is one of the important measures that are observed in the ini- tial models. The period time and how it changes is observed in each analysis. This parameter can be calculated straight from building standards but the calculations will be on a highly advanced level when including the piling design. An important feature when designing buildings here in Gothenburg is to incorporate the piling deign. The piles has an important role with their sti�ness and are modelled to- gether with the structural system in ETABS. When designing a high-rise building, the sti�ness is usually the parameter deciding the final design. But when we start to construct higher buildings, as the Karlator- net, exceeding the 200m mark, the acceleration will instead govern the design of the structural system, Adreas declares. The FEM model is created with predicted sti�ness, for beams connecting the core and the bottom slab, sti�ness values along with the degree of concrete cracking are based on experience. When a structural system is decided upon, VBK is quick to incorporate a wind consult into the process. The wind consults will analyse the building proposal and generate an early assessment of the wind loads based on their knowledge and ex- perience. These wind loads received from an early assessment are used in further design of the building until a wind tunnel test is performed. Regarding the skyscraper Karlatornet, the exposing factors were not decreased as o�ensive as in other projects since the building is placed considerable close to the ocean where stronger winds occur compared to other buildings constructed in the city. The structural system of Karlatornet include outriggers that also were analysed in an early FEM model to decide upon at which height to put them and how it will a�ect the frequency. Before a final wind tunnel test were performed of Karlatornet, a workshop in wind tunnel design was conducted. In this workshop, which is a huge investment for the design of a high-rise building, several iterations of the building design were made and analysed. Di�erent locations, angles, designs, heights etc. were tested in the workshop before a final solution were chosen. A final wind tunnel test generate wind loads in horizontal x- and y-direction as well 21 3. Interview study as the rotation. These loads are specified for each individual floor and then com- bined in 24 di�erent load combinations. The wind loads are based on the current topography as well as future additional structures predicted to surround the build- ing. The wind loads are observed in both service limit state and ultimate limit state. The natural frequency will vary depending on observed limit state. In ultimate limit state the concrete is cracked and subsequently weaker. At the same time the damp- ing will be increased at ultimate limit state, compared to service limit state, which also a�ects the wind load. To sum up, the design is based on 96 di�erent wind loads; current topography, future topography, SLS and ULS combined in 24 load combinations. Besides the most common wind tunnel test HFFB, Karlatornet were tested for cladding pressure, HFPI, where the local pressure on the facade is measured by adding several taps on the model. Andreas explains that by a rough estimation, the facade cover an equal large investment cost as the core itself. Subsequently it is possible to lower the cost considerably by observing the local wind pressure at the facade and design accordingly. In Karlatornet, the local wind pressure is greater at the twist as well as at building corners and top, depending on building side ob- served. Adapting the facade depending on local wind pressure saved the project a large amount of investment cost. In the project of Karlatornet, there have been a detailed consideration regarding di�erent directions of wind velocity. The wind tunnel test is based on directional dependent wind data where the wind load is higher from southwest to northwest. The resulting wind loads for each direction are displayed in a so called wind rose and the building is designed accordingly. Gathered results from wind tunnel tests shows significant decrease in wind loads compared to prediction based on building standards. Andreas speaks of how it is a complex task to conclude any acquaintances between a wind tunnel test and wind loads by building codes to define any guidelines to use in future projects. 3.2.2 A new method for design The cost of a wind consulting early assessment is rather low, around 50 000 SEK. It is a rather cheap investment in relation to its favorable outcome, Andreas express. Of this reason, a new method for early design may not be worth its time. Wind consults are considerable competent and exceptional knowledge would be required to reach their results for predicting wind loads in an early stage of high-rise building design. However, further guidelines for the design of high-rise buildings that would bring the engineer even closer to the "true" solution would certainly strengthen the design concept, especially in an early stage. Some buildings are of complex manners, for example Region City here in Gothen- burg. The buildings are really close to each other which will create di�erent wind 22 3. Interview study movement and pressure on the structures. Additionally, the buildings share foun- dation which will challenge the design further since the wind loads are taken care of at the foundation. Even though one can predict that heavy wind gusts will not occur at the same time and with the same direction on the buildings, it is still a complicated task to perform without a wind consult in charge. In a case like this, the design will be based on di�erent terrain categories for various directions facing the buildings’ all sides as well as a complex wind movement between the buildings. Andreas explains that this kind of analysis is way beyond engineering practise and needs to be addressed by a wind specialist. One disadvantage of including a wind consult in an early phase of the project is the requirement of making a suitable choice of wind consult company to work with. Even though the main idea with the actual wind tunnel test is to confirm and re- fine the results from the early assessment, the project is not closed to working with the same wind consultant as in the early stage. The choice of wind design team is based on degree of proficient but also important, a matter of investment cost. Subsequently, changing company could be misleading when comparing wind tunnel tests to an early assessment of wind loads. An interesting observation Andreas proposed was to bring forth more clear guide- lines regarding the total displacement at the top of the building. Today, there are only recommendations exsisting without any context to relevance for the design of high-rise buildings in particular. Since the slenderness of the building have an sub- stantial a�ect on the predicted wind load, a more slender design will generate higher wind loads, the drift is an vital feature in design. In reality the wind loads will be lower at parts of the building where the sti�ness is higher, but in design by building standards, these dynamic wind e�ects is kept constant. The total horizontal defor- mation at the top of the building is measured against di�erent ratios in the early FEM analysis and often chosen to match a recommendation of h/500. If one were to be more comfortable of choosing a higher displacement, like h/400, this would generate a whole other building design, Andreas expresses. Of course this kind of guidelines must include the consequences of accepting a higher displacement value such as shear deformations that a�ects the design of the facade. It would be interesting to gather results from wind tunnel tests for buildings in dif- ferent ranges of height, for example 50-100m, 100-150m, 150-200m etc and compare relevant reduction factors with the Eurocode building standards, Andreas declares. Connecting building height with relevant reduction factors would be a profitable development of current guidelines in design. To bring even more knowledge in a method like this, the frequency could be an additional parameter to observe. 23 3. Interview study 3.3 Interview II - Marco Binfaré The second interview is with Marco Binfaré who is a building designer at WSP in Sundsvall. Reference buildings for this interview is Norra Tornen, located in Vasastaden, Stockholm, see Figure 3.2. Norra Tornen consists of two buildings, Helix of 111 meters in height and Innovationen of 120 meters in height. The buildings’ primary use is residential. The interview with Marco was performed via a Teams meeting. Figure 3.2: Norra tornen in Vasastaden, Stockholm. 3.3.1 Design of high-rise buildings The design of high-rise buildings at WSP is an iterative process together with the architects for a certain project. For each architectural proposal, a simple analyze of the buildings’ stability is conducted. In this investigation the buildings’ slenderness is observed in relation to the location of the core. The hand-made calculations are made in a rough and general matter, sometimes random detail checks are made, but no detailed calculations is conducted in this early stage. When a final layout of the structural system is chosen together with the architects, a more detailed anal- ysis of the building is conducted using FEM modelling. The whole development of high-rise building design is an iterative process, where changes can be necessary even in a later stage. Marco explains how the iteration of FEM models often results in a variety of approximately 10 models with di�erent properties. The most com- mon FEM-program used at WSP is FEM-Design but Marco also talks about the 24 3. Interview study advantage of using the FEM program ETABS particularly for the design of high-rise buildings. In the beginning of the design process, wind loads are conducted using the Eurocode European standards since they give more conservative and higher values. The load cases are observed in both service limit state as well as ultimate limit state. One of the most important parameter for the design of high-rise buildings is the natural frequency, Marco explains. A desirable value of the natural frequency is approxi- mately 0.35 Hz. An important matter is to avoid to get the torsion mode as first mode, favorable would be to get the x- and y-component as first two modes and subsequently torsion as last and third mode. This is due to the di�culty to control the torsion of the building. Subsequently looking at the natural frequency is a useful tool to analyze the buildings’ design. This will tell if the structural system is in need of further core walls for stability. Demands of the natural frequency is missing in to- day’s building codes, there is only a few recommendations to be found. The natural frequency of the building will a�ect the acceleration at the top of the building and is therefore an vital parameter to observe. Demands on acceleration for high-rise buildings are also neglected in today’s standards, although ISO standards do have some inputs, there is a lack of specified demands. The acceleration is observed after a chosen layout is set and will determine whether the building will be suitable for residential or o�ce use. Another feature to observe in the design is the slenderness of the building. Marco explains that there is a lot of studies about this conducted in the US, but Eurocode lack the same level of guidelines. Often in design, the choice of total horizontal deflection at the top of the building is based on experience from previous projects. In later design of high-rise buildings, a wind tunnel test is performed. There is no early assessment conducted before the actual wind tunnel test. If there is a need to investigate a special type of facade that is not covered in the Eurocode, a CFD analysis can be implemented before the wind tunnel test. In general, one wind tun- nel test is performed per project. Although, if there are some uncertainties of the resulting wind loads, for exapmle if there is a significant decrease of wind loads com- pared to the analysis by building standrads, a second opinion could be conducted. The resulting wind loads from the wind tunnel test are approximately 15-20% lower than the wind loads conducted by building standards. Depending on the type of core used in design it will induce di�erent consequences. For a core made of steel, a 20% decrease in wind loads will have a huge impact, while the same reduction of wind loads for a concrete core may not give a significant di�erence. Although, the reduction of wind loads will nevertheless save material usage and investment cost together with possible improvements of frequency and acceleration. From the wind tunnel test, the dominant wind directions are displayed by a wind rose. It demonstrate the most critical load case of the side of the building facing the dominant wind direction, which is dependent on the unique design of the building. With the dominant wind direction in mind, 40-50 di�erent load cases in 24 load combinations are conducted and combined to find the most critical case. 25 3. Interview study For the project of Norra Tornen, a cladding wind tunnel test, HFPI, was performed since the buildings’ facade are of a special design. In this analysis there is an detailed observation of attachments and fine technicalities. The cladding analysis that gener- ates the local pressures on the buildings’ facade resulted in an partition of segments that could be designed with di�erent wind loads. By implementing this analysis, the structure reaches its optimization. Although, one do not want do optimize to the maximum, Marco explains, especially for a project like this where the core is made out of prefabricated concrete elements due to possible mounting di�culties. A challenge in the project of Norra Tornen, Marco explains, was the non-centred core. The core was instead located markedly to the right, which gave rise to a rotational di�culty. The two buildings had some disparity to each other. The Helix building was in need of an additional core wall in the outer part of the floor layout since it is more slender than the other building. This was a case where the engineer got involved later than the architects and the design resulted in a need of additional stability. Favorably, the engineers should be involved in the early phase to minimize these kinds of late changes in design. Furthermore, the project Norra Tornen brought challenges with its prefabricated core which gave rise to problematic joints and an additional demands in transferring the loads through the structure to gain stability. The buildings are designed with overhangs of the facade in three di�erent directions with each a length of five meters each whereupon the structure rise even 10 floors higher. Marco describe the project as a fun but long process. 3.3.2 A new method for design There is a big lack of guidelines and methods in today’s European building codes re- garding the design of high-rise buildings. There is a lot of literature to gain from the US, Canada and UK regarding high-rise buildings. Marco explains that the reason for this inadequacy in the European standards could be of the simple reason that there has not been a need for designing high-rise buildings, until now. Especially here in Sweden, there is not a lot of tall buildings and especially not close to the height of international skyscrapers. According to Marco, guidelines or a manual for designing high-rise buildings to- gether with a detailed literature collection would be desirable. To develop a method in early design to assess if the core of the building is adequate or not, would benefit the design process. A method or guidelines of this kind could include principally key numbers to follow and include an amount of three or four bullet points to be verified. Another important aspect for the design of high-rise buildings that Marco declare is to consider the consequence of carbon dioxide. This kinds of tall buildings that could amount approximately 400 concrete walls with high concentration of cement are not to be ignored. This is a component in the design of high-rise buildings that can and should be optimized for the environments’ sake. It should be a part of the early design where depending on the span and slab, an analyse of the amount of 26 3. Interview study carbon dioxide emission is conducted. The construction of tall buildings has only begun and there will be more, Marco express. There is an opportunity to optimize the use of concrete in high-rise buildings, especially when observing the slabs. At WSP the consideration of carbon dioxide emissions in an early stage has not been conducted in the past. There is a current development of method regarding net zero at WSP in the UK. In this method, one can adjust the amount of floors, material, spans and variation of slab to observe the expected carbon dioxide emission. There is a lot to gain by this kind of analysis, Marco explains. 27 3. Interview study 3.4 Interview III - Gustav Söderlund This interview is performed with Gustav Söderlund who is a building designer at PE Teknik & Arkitektur in Gothenburg. Gustav is one of the designers behind Hotel Draken which is a high-rise building of 100 meters in height, located in Gothenburg, see Figure 3.3. The building which is a hotel was constructed between the years 2020 to 2023. The interview with Gustav was performed at PE’s o�ce in Gothenburg. Figure 3.3: Draken Live at Järntorget, Gothenburg. 3.4.1 Design of high-rise buildings At PE Teknik & Arkitektur, the design of high-rise buildings is dominated by im- plementing FE modelling in an early stage. Analysing the building by using FEM modelling is an easy and quick way of design compared to using hand calculations, Gustav explains. The FEM program that Gustav prefers for high-rise buildings is ETABS. ETABS is a program with more options compared to other FEM programs such as boundary conditions and several functions favorable for high-rise buildings. It is not that important to go into details in an early stage of modelling, it is more about analysing approximately how large the core needs to be. In the beginning the model is a general layout that is copied and added on each floor of the building to get an overview of the structural systems’ response. In the early stages of design, the location of the columns is not crucial for the sta- bility. The main task is to find a structural system that falls between the core walls that has been conducted together with an architectural idea of layout. It is an itera- tive process between the architectural design, demands on accessibility and technical spaces and the structural stability. The structural system is adjusted repeatedly by either increasing the thickness of the core walls or by modifying the geometry. In 28 3. Interview study some special cases, there could also be a need of integrating an outrigger in design, especially for higher and more slender buildings. The wind loads are included in the model by using Eurocode European standards. Gustav disclose that when using the Eurocode standards there can be a huge dif- ference to the wind loads gained from a wind tunnel test. In an early project stage, there is an early assessment conducted by wind specialists. After adjusting the design to match their recommendations, an actual wind tunnel test is performed. Observing the results from the wind tunnel test, the wind loads in the project Draken were decreased with approximately 20% compared to the loads generated from cal- culations by building standards. Additionally, there is a dynamic factor that needs to be included in the design of high-rise buildings. This dynamic factor tends to be of a high value when generated by hand calculations, while the wind experts can decrease this value based on their experience of the buildings’ structure, location and the known behaviour of the wind. At PE, Gustav explains that they tend to lower the value a bit in the beginning, but always lean against the word of the hired wind specialists to get a professional opinion. At PE Teknik & Arkitektur the wind tunnel test used for projects of high-rise buildings has been the high-frequency force balance (HFFB) kind of wind tunnel test. Except for the wind loads itself, accelerations are a crucial issue for high-rise build- ings. The acceleration needs to be limited to a certain level suitable for the buildings’ intended use. In this matter, Gustav points out that we as engineers are fixed to the current building standards, while the wind consults possess greater experience regarding building dynamics. The acceleration is depending on the buildings’ mass and subsequently its inertia. For a lightweight building, accelerations at the top could be a problem. Gustav explains that his interpretation of acceleration for buildings is that it starts to be a problem when exceeding 100m in height and prob- ably escalates from there. When looking at the acceleration demands, these di�er between the Eurocode Eu- ropean standards and the EKS, where the values retrieved from the Eurocode is lower. Even though this is a one parameter variety, the result will di�er with ap- proximately 30%. Gustav then describes that is has been a challenge to compare these accelerations generated by building standards to the accelerations gained from the wind consults. Looking at the results generated by wind specialists, the damping has been set to 15% which is not the case when following current building standards. In the project Draken, Gustav talks about the challenge in integrating acceleration together with piling design. When the piles are deforming, it creates a weakness of the structural system. Normally, some piles are placed diagonally in the ground to be able to take horizontal loading. In this project, the piles were vertically placed, which created another demanding task. The limitation of acceleration chosen for the project Draken was a value between the recommendations of a residential and o�ce used building. 29 3. Interview study Although service limit state dominates the design, ultimate limit state is investi- gated for the matter of tension. If any adjustment is needed in regards to the ultimate limit state, it is normally solved by changing to a better concrete qual- ity or adapting the reinforcement. Issues in design for ultimate limit state is not a geometrical concern as it is for the service limit state verification. As of now, the concrete weight is not being analyzed in projects at PE Teknik & Arkitektur, Gustav declares. Regarding the total deflection at the top of the building, Gustav explains that the most common choice is by following the recommendation of h/500. Another challenge in the project of Draken was the prefabricated elements and how to handle the joints and connections that occur. The idea were to make the core act as a whole homogeneous structure regardless its segmented division. To make this happen it was important that the connections were sti� enough. This is an issue that do not occur when constructing in-situ, where the task is more or less to only reinforce correctly. 3.4.2 A new method for design Looking at the 10% di�erence between wind loads gained from calculations based on building standards and wind loads received from wind tunnel tests, there is defi- nitely room for development of guidelines for a early design assessment of high-rise buildings. But at the same time, wind is a complex phenomena, Gustav declares, and require a lot of information to analyse properly. It would be useful to create a method to secure demands on accelerations for high- rise buildings, Gustav express. Both Eurocode European standards and the EKS have methods for calculating the acceleration where the Eurocode measurements usually generate a lower value of acceleration. When following the EKS guide, a re- turn period of five years is implemented while the return period in Eurocode is one year. Gustav explains that the factors received for di�erent types of return periods are di�erent depending on source. As Gustav further mentions, the accelerations re- ceived from the wind consultants are sometimes di�cult to match against the values gained from calculations by norm. There is no easy way to find a common approach to use in future projects before including a wind specialist. Final words about the importance of acceleration, Gustav talks about how it is crucial to solve this issue early. There is no way to improve the concrete quality for increasing the bending sti�ness. In today’s building industry there are more and more common to construct timber buildings with lower self weight compared to concrete or steel structures. It will then be even more vital to handle the ac- celerations since it is mainly the mass and inertia that decreases the acceleration. When observing concrete buildings, a structure of 100 meters usually do not endure problems regarding accelerations. The di�culty starts beyond this, when there is a need of outriggers and specialized wind analyses. 30 4 Case study A case study is implemented in this thesis to support the development of guidelines for design of high-rise buildings in an early stage. To conduct a comparison between building standards and wind tunnel tests, this thesis has been provided with three di�erent reference buildings with appurtenant data. With respect to the building designers, the three di�erent buildings are presented anonymously with titles; Build- ing A-C. Due to this anonymous commitment, limited information of the buildings are presented in this chapter. The data provided focus on the buildings’ structural systems along with wind properties and wind tunnel testing results. The buildings analysed are of di�erent heights and configuration to bring a variety of analysis. 4.1 Building A Building A is a 111 meters high building with 34 floors. The buildings length is decreased with its height where the largest length is of 48 meters while the smallest section is only 22 meters. Architectural drawings of Building A is shown in Figure 4.1 to 4.3. The building are mainly of residential usage and is located where a fundamental wind velocity of 24 m/s is assumed. Structural properties of Building A along with wind properties are presented in Table 4.1 and 4.2. Table 4.1: Structural properties for Building A. Building height 111 m Building minimum width 15 m Building length 22-48 m Slenderness ratio 7.4/1 Across-wind dominance 0.164 < 1/3 Shear wall thickness 400 mm Slab thickness 200 mm Table 4.2: Wind load properties for Building A. Terrain category II Basic wind velocity 24 m/s Peak wind velocity pressure 1.37 kN/m2 Equivalent mass me 193332 kg 31 4. Case study Figure 4.1: Entrance floor plan of Building A. Figure 4.2: Rooftop floor plan of Building A. 32 4. Case study Figure 4.3: Section A-A respectively section B-B of Building A. To design Building A against strong winds and optimize its structural system, a wind tunnel test was performed for both structural and cladding design. Gust-induced vibrations as well as along- and across-wind accelerations were investigated. Natural frequencies for mode 1 to 3 from wind tunnel testing are presented in Table 4.3. The wind tunnel test was conducted with 36 di�erent wind directions in steps of 10 degrees (0-350). The building was modelled both as solitary configuration as well as with close buildings within a radius of 250 m. The model in the wind tunnel test was on a scale model of 1:250 with approximately 3200 pressure taps covering the facades. In this thesis, the dominant wind direction is analysed and subsequently compared to wind tunnel test results on the same side. Table 4.3: Frequency for mode 1 to 3 for Building A. Mode Frequency Mode 1 0.319 Hz Mode 2 0.470 Hz Mode 3 0.550 Hz 33 4. Case study 4.2 Building B Building B is a 94.7 meters high circular shaped building with 29 floors. The build- ing is of residential usage and located in a terrain that is characterized as type II. The building is designed with a 24 cornered facade, which in building code regula- tions is designed as a circular structure. The circular geometry induce wind issues such as increased formation of vortex shedding. Further on, the fundamental wind velocity is assumed to be 24 m/s. Structural- and wind load properties of Building B is presented in Table 4.4 and 4.5 followed by architectural drawings in Figure 4.4 to 4.6. Table 4.4: Structural properties Building B. Building height 94,7 m Building minimum width 25 m Building length 51 m Slenderness ratio 3.8/1 Along-wind dominance 0.54 > 1/3 Shear wall thickness 400 mm Slab thickn