1

CHAPTER I

INTRODUCTION

1.1 GENERAL

Steel frame buildings with perimeter special moment resisting frames (SMFs)
designed in highly seismic urban regions in Chennai, should comply with regional seismic
design code requirements (IS 1893 part 1 2016, IS 800 2007). Normally, deep beams are
used in SMFs in order to control the lateral drift demands. Therefore, designers tend to use
deep wide-flange steel sections as columns to satisfy the strong-column-weak-beam
criterion. These sections are able to respect economy due to their low weight-to-inertia
ratio. The term “deep” refers to wide-flange sections with depth larger than 400mm (i.e.,
16 inches). The web and flange local slenderness ratios of deep wide-flange steel columns
comply with the seismic compactness limits applied to highly ductile members as discussed
in IS 800 2007 However, deep wide-flange sections with high web and flange slenderness
ratios (i.e., close to the compact limit for highly ductile members, λhd ) are susceptible to
local and lateral torsional buckling instabilities. As part of a first-story SMF, steel columns
typically experience large lateral drifts combined with high compressive axial load levels
during a severe seismic event. As a result, it is expected that once plastification occurs at
the base of these columns, they could rapidly deteriorate in flexural stiffness and strength
and possibly trigger dynamic collapse of a steel SMF. Typically, the first few stories of a
SMF form a global collapse mechanism due to P-Delta.

This paper investigates the cyclic behavior of wide-flange steel beam-columns through
detailed FE analysis. A finite element modelling approach is first proposed to model steel
beam-columns. The FE approach is then validated with past experimental studies. The
cyclic behavior of a comprehensive set of wide-flange steel sections typically used as steel
columns in steel construction practice of steel SMFs is investigated analytically. Emphasis
is placed on the cyclic behavior of interior steel columns in which the axial load is almost
constant during an earthquake (i.e., negligible effects due to dynamic overturning
 2

moments). Thus, each steel column is subjected to a symmetric cyclic lateral loading
protocol combined with different levels of constant compressive axial load ratios. In
particular, the effect of cyclic deterioration n the beam-column flexural strength is
assessed. The expected pre-capping plastic rotation and the corresponding plastic hinge
length of a steel beam-column are quantified depending on the level of the applied axial
load ratio. The pre-capping plastic rotation predicted by the FE model is used to assess the
current analytical modelling recommendations for steel columns as per Steel Table.
Finally, the level of axial shortening that the wide-flange steel beam-columns experience
under different axial load levels is quantified.

1.2 STRUCTURAL STEEL

Structural steel is a standard construction material, made from specific grades of
steel and formed in a range of industry standard cross-sectional shapes. Structural steel
grades are designed with specific chemical compositions and mechanical properties
formulated for particular applications.

In most countries Structural Steel is regulated and must meet a minimum specific
criterion for Shape, Size, Chemical Composition, strength etc. Chemical Composition of
Structural Steels – E165, E170 & E215

It is a fundamental factor which defines the Mechanical properties of the steel

material. In the following table you can see the Max % levels of certain regulated elements
present in Indian Structural steel grades E165, E170 & E215.

Table 1.1 Chemical Composition

Grade Designation C Mn S P

E165 0.25 1.25 0.060 0.075

E170 0.25 1.25 0.06 0.075

E215 0.25 1.25 0.06 0.075

 3

The product analysis shall be carried out on the finished product from the standard
position. Permissible limits of variation in case of product analysis from the limits specified
in Table 1.2 shall be as given below.
TABLE 1.2 Product composition limits

Permissible Variation Over the

[Link] Constituent
Specified Limit, Present, Max

1 Carbon 0.02
2 Manganese 0.05
3 Silicon 0.03
4 Copper 0.03
5 Sulphur 0.005
6 Phosphorus 0.005
1.3 TENSILE TEST

Tensile strength, yield strength and percentage elongation when determined in Accordance
with IS 1608 shall be as given in Table 1.3.
Table 1.3 Mechanical Properties of steel
Grade Tensile Yield Percentage Internal
Designation Strength Stress, Elongation, A Bend
Rm ReH at Gauge Diameter
Min Min Length, Lo Min
MPa MPa 5.65√So Min
E 165 290 165 23 2t
E 170 330 170 23 3t
E 215 370 215 23 3t
t is the thickness of the test piece. 1 MPa = 1N/mm2 = 1MN/m2 = 0.102 kg f/mm2 = 144.4
psi.
 4

1.4 SCOPE AND OBJECTIVES

To provide the basis for performance evaluation of Beam and columns in braced frames
subjected to Cyclic Loading actions,

the objectives of this research are to:

(1) Develop a statistical and Dynamic Lateral load-based loading sequence for
Analytical investigation on beam-columns,
(2) Determination of Axial shortening · Strength deterioration · Pre-capping plastic
rotation.
(3) Determination Column plastic hinge length
 5

CHAPTER II

LITERATURE REVIEW

2.1 GENERAL

An extensive literature review was carried out prior to the project. The survey of
literature includes various types of Cyclic Behavior, Connection design in Different Types
of steel.

2.2 LITERATURE REVIEW

James Newell, Chia-Ming Uang (2006),“Cyclic Behavior Of Steel Columns With

Combined High Axial Load And Drift Demand” Nonlinear time-history analysis of steel
braced frames. Buckling restrained braced frames (BRBF), subjected to earthquake ground
motions has revealed that columns in the bottom stories are often subjected to combined
high axial load and inelastic rotation demand resulting from story drift. Nine W14 section
column specimens have been subjected to laboratory and analytical investigation.3-story
and 7-story BRBF prototype building models were designed and analyzed. Nonlinear time-
history analysis of frame models, subjected to a suite of 20 properly scaled earthquake
ground motions, was conducted. Column specimens were tested at the University of
California, San Diego (UCSD) Seismic Response Modification Device (SRMD) Test
Facility. The test data agreed well with the established P-M interaction surface. Specimens
achieved drift capacities of 7% to 9%, corresponding to a ductility of approximately 10.
These drift capacities were more than four times the expected BRBF story drift from
earthquake nonlinear time-history analysis.

Emanuele Grimolizzi, Stefan Calogero Cravotta(2015), “Simulation Of Vehicle

Impact Into A Steel Building” Understanding the true behavior of impacted structures is
the only way to assess their robustness under exceptional events such as vehicle collision.
The primary objective of this master’s thesis was to perform a finite element parametric
investigation on the influence that some parameters have in steel buildings subjected to
vehicle impacts. By using the Abaqus software, a finite element model of the structure has
 6

been created. The five-story steel building considered has been modelled in a simplified
manner with the exception of the impacted area which, instead, has been defined in a more
detailed fashion. During the simulations, different preload conditions have been used,
comparing cases with and without the preload force. Regardless its variation, it has not
been observed any increase in the structural resistance. For all the cases considered, an
increase of the material ductility, achieved by increasing the ultimate strain at failure,
entailed higher resistance of the connections. Various failure modes have been observed
when the material properties have been changed.

Bhavana B, Abhishek N Naik(Oct 2017),Analytical Investigation Of Thermal

Behavior Of Steel Structural Beam Component” The architectural flexibility of steel
structure dominates over other conventional structures, but fire and its associative load are
one of the major accidental loads, which could be fatal in steel structures. Load bearing
capacity of steel structural components drastically reduces in the fire condition. In this
paper, the performance of steel structural components under accidental fire loads is
investigated. Constraints in the experimental investigation lead to the analytical
investigation by the method of finite elements. ISO 834 is adopted as standard fire curves
for modelling up-to temperature of 6800 C. Computational structural models including the
earthquake loads will be analyzed with the help of advanced structural software. Beam
components of the critical section are optimized and processed to thermal analysis. In
general effects of rising of temperature in the fire hour induce an expansion in beam
components. If the expansion is restrained, stress induces over the region of restraining,
resulting in the change or rise in deformation. In this attempt, the study of behavior on non-
coated and protective coated steel beam structures under direct thermal loading with
relation with the total deformation and stress- strain are investigated with help of FEM
based software.
Nimmi K. P, Krishnachandran V.N (Aug 2016), “Experimental and Analytical
Investigations of Cellular Steel Beams” Their design and making process provides
greater flexibility in beam proportioning for strength, depth, size and location of holes. The
 7

advantages of manufacturing these beams is to increase overall beam depth, the moment
of inertia and section modulus, which results in greater strength and rigidity. Cellular
beams are the castellated beams with circular openings. Cellular beams are used as primary
or secondary floor beams in order to achieve long spans and service integration. They are
also used as roof beams, and are the best solution for curved roof applications, combining
weight savings with a low-cost manufacturing process. The purpose of the current research
is to study optimum shape and size, stiffness and load carrying capacity under applied load
and finite element analysis of non-composite castellated beams. The first part of the
research program focuses on the comparison of experimental and analytical study to
optimize the shape of openings and verify the accuracy of finite element analysis on
ANSYS software. In the second part of the research, parametric study conducted to
optimize the parameters effecting the structural performance of cellular beams by finite
element analysis. The parameters adopted in this work are consistent with BS-5950.

A.A. El Damatty , M. Abushagur, M. A. Youssef (2002), “Experimental and

Analytical Investigation of Steel Beams Rehabilitated Using GFRP Sheets” The
current study investigates a methodology to enhance the flexural capacity of steel beams
by bonding Glass Fiber Reinforced Plastic (GFRP) sheets to their flanges. A heavy-duty
adhesive, tested in a previous study is used to bond the steel and the GFRP sheet. In
addition to its ease of application, the GFRP sheet provides a protective layer that prevents
future corrosion of the steel section. The study reports the results of bending tests
conducted on a W-shaped steel beam before and after rehabilitation using GFRP sheets.
Enhancement in the moment capacity of the beam due to bonding GFRP sheet is
determined from the test results. A closed form analytical model that can predict the yield
moment as well as the stresses induced in the adhesive and the GFRP sheets of rehabilitated
steel beam is developed. A detailed finite element analysis for the tested specimens is also
conducted in this paper. The steel web and flanges as well as the GFRP sheets are simulated
using three-dimensional brick elements. The shear and peel stiffness of the adhesive are
modeled as equivalent linear spring systems. The analytical and experimental results
 8

indicate that a significant enhancement in the ultimate capacity of the steel beam is
achieved using the proposed technique. The finite element analysis is employed to describe
in detail the profile of stresses and strains that develop in the rehabilitated steel beam.

Emanuele Grimolizzi, Stefan Calogero Cravotta (2015), “Simulation Of Vehicle

Impact Into A Steel Building” Understanding the true behavior of impacted structures is
the only way to assess their robustness under exceptional events such as vehicle collision.
The primary objective of this master’s thesis was to perform a finite element parametric
investigation on the influence that some parameters have in steel buildings subjected to
vehicle impacts. The parameters chosen for the study, involved uncertainties in the material
definition and in the load configuration of the bolts used in the impacted column end-
connections. By using the Abaqus software, a finite element model of the structure has
been created. The five storey steel building considered has been modelled in a simplified
manner with the exception of the impacted area which, instead, has been defined in a more
detailed fashion. During the simulations, different preload conditions have been used,
comparing cases with and without the preload force. Regardless its variation, it has not
been observed any increase in the structural resistance. On the other hand, the simulation
provided interesting results for what concerns the material variations in the bolts. Although
the changes have been small in magnitude, the effect on the structural response during the
impact was remarkable. For all the cases considered, an increase of the material ductility,
achieved by increasing the ultimate strain at failure, entailed higher resistance of the
connections. Various failure modes have been observed when the material properties have
been changed. Having clarified the influence of the assumptions made, the results provided
helpful information in sight of future studies. Although the model still needs to be
validated, the research clarified which of the parameters investigated are to be collected
with more attention.
 9

K. Al Fakih, S. C. Chin and S. I. Doh (Feb 2017), “Finite Element Analysis of Beam
to Column Bolted Connection – A Review” This paper contains the review of studies
performed on the behavior of steel beam-to-column connections by using finite element
analysis computer package known as “Abaqus, Ansys, etc” from the theoretical and
practical points of view. The objective of this article is providing the foundation for the
design and analysis faster and more economical and securing the required strength for the
connection steel. Methods/Statistical Analysis: Nearly seventy-one academic and popular
research/literature in the field of steel connections had been studied in which an overview
of recent events has made the construction of the system viable in the infrastructure.
Findings: Linear analysis of connection behavior is very complicated since there is no
immediate solution for it. All the results of the accounts of the design connections cannot
be verified only through lab tests which consume a lot of time, money and effort. Therefore,
the availability of simulation programs can overcome these problems. This article will be
of value to anyone seeking better understanding in the area of steel beam-to-column
connections behavior. Application/Improvements: Review study presented in this paper
can be used as reference for future investigation especially in the member strength and
development of the design approach for steel beam-to-column connections with simulation
models.
 10

CHAPTER III

METHODOLOGY

FINITE ELEMENT MODELING AND METHODOLOGY

3.1 INTRODUCTION

The analysis of stress and deformation of the loading of simple geometric structures
can usually be accomplished by closed-form techniques. As the structures become more
complex, the analyst is forced to use approximations of closed-form solutions,
experimentation, or numerical methods. There are a great many numerical techniques used
in engineering applications for which digital computers are very useful. In the field of
structural analysis, the numerical techniques generally employ a method which discretizes
the continuum of the structural system into a finite collection of points (or nodes)/ elements
called finite elements. The most popular technique used currently is the finite element
method (FEM). Other methods some of which FEM is based upon include trial functions
via variational methods and weighted residuals the finite difference method (FDM),
structural analogues, and the boundary element method (BEM).

3.2 THE FINITE DIFFERENCE METHOD

In the field of structural analysis, one of the earliest procedures for the numerical
solutions of the governing differential equations of stressed continuous solid bodies was
the finite difference method. In the finite difference approximation of differential
equations, the derivatives in the equations are replaced by difference quotients of the values
of the dependent variables at discrete mesh points of the domain. After the equations are
replaced by difference quotients of the values of the dependent variables at discrete mesh
points of the domain. After imposing the appropriate boundary conditions on the structure,
the discrete equations are solved obtaining the values of the variables at mesh points. The
technique has many disadvantages, including inaccuracies of the derivatives of the
approximated solution, difficulties in imposing boundary conditions along curved
 11

boundaries, difficulties in accurately representing complex geometric domains, and the

inability to utilize non-uniform and non-rectangular meshes.

3.3 THE BOUNDARY ELEMENT METHOD

The boundary element method developed more recently than FEM, transforms the
governing differential equations and boundary conditions into integral equations, which
are converted to contain surface integrals. Because only surface integrals remain, surface
elements are used to perform the required integrations. This is the main advantage of BEM
over FEM, which require three-dimensional elements throughout the volumetric domain.
Boundary elements for a general three-dimensional solid are quadrilateral or triangular
surface elements covering the surface area of the component. For two-dimensional and
axisymmetric problems, only line elements tracing the outline of the component are
necessary. Although BEM offers some modelling advantages over FEM, the latter can
analyze more types of engineering applications and is much more firmly entrenched in
today’s computer-aided-design (CAD) environment. Development of engineering
applications of BEM is proceeding however, and more will be seen of the method in the
future.

3.4 FINITE ELEMENT METHOD

The finite element method (FEM) is a numerical procedure for analyzing structures and
continuum. The problem may concern to obtain approximate solutions to solid mechanics
heat transfer and fluid mechanics problem. Basic ideas of the finite element originated from
advances in aircraft structural analysis. In 1941 Hrenikoff presented a solution of elasticity
problems using frame work method. Couroint, in 1943 used piecewise polynomial
interpolation over triangular sub regions to model torsional problems. A book by Argyris
in 1955 on energy theorems and Matrix methods laid a foundation for further developments
in finite element studies. Turner et al. derived stiffness matrices for truss, beam & other
elements & presented in 1956. The term finite element method was first coined and used
by Clough in 1960. The first book on finite elements by Zienkiewicz and Chung was
 12

published 1967. The stress analysis in field of Civil, Mechanical and Aerospace
engineering, Naval architecture, Off shore engineering and Nuclear engineering is
invariably complex and for may of the problems it is extremely difficult and tedious to
obtain analytical solutions. In these solutions, engineers usually resort to numerical
methods to solve problems. With the advent of computers, one of the most powerful
techniques that have been developed in the realm of engineering analysis is the finite
element method. The finite element method can be used for analysis of structures/solids of
complex shapes and complicated boundary conditions. Its applications range from
deformation and stress analysis of automotive, air craft, building and bridge structures to
field analysis of heat flux, fluid flow, magnetic flux, seepage and other flow problems.
With the advances in computer technology and CAD systems, complex problems can be
modelled with relative ease; several configurations can be tested on computer before the
first prototype is built.

3.5 BASIC CONCEPT

In the finite element method of analysis, a complex region defining a continuum is

discretized into simple geometric shapes called finite elements. The material properties and
the governing relationships are considered over these elements and expressed in terms of
unknown values at elements corners. An assembly process duly considering the loading
and constraints results in a set of equations. Solution of these equations gives the
approximate behavior of the continuum.

3.6 BASIC STEPS IN THE FINITE ELEMENT METHOD

The following are the steps adopted for analyzing a structural engineering problem by the
finite element method.

3.6.1. DISCRETIZATION OF THE DOMAIN

The continuum is divided into a number of finite elements by imaginary lines or surfaces.
The interconnected elements may have different sizes and shapes. The choice of the simple
 13

elements or higher order element straight or curved, its shape, refinement are to be decided
before the mathematics formulation starts.

3.6.2. IDENTIFICATION OF VARIABLES

The elements are assumed to be connected at their intersecting points referred to as

nodal points. At each node, generalized displacements are the unknown degrees of
freedom. They are dependent on the problem at hand. For example in a plane stress problem
the unknowns are two linear translations at each nodal point.

3.6.3. CHOICE OF APPROXIMATING FUNCTIONS

Once the variables and local coordinate system have been chosen. The next step is the
choice of displacement function. In fact it is the displacement function that is the starting
point of the mathematical analysis. This function represents the variation of the
displacements within the element. The function can be approximated in a number of ways.
The displacement function may be approximated in the form of a linear function or a higher
order function. The shape of element or the geometry may also be approximated. The co
ordinates of corner nodes define the element shape accurately if the element is actually
made of straight line or plates.

3.6.4. FORMATION OF THE ELEMENT STIFFNESS MATRIX

After the continuum is discretized with desired element shapes, the element stiffness
matrix is formulated. This can be done in a number of ways. Basically, it is a minimization
procedure whatever may be the approach adopted. For certain elements, the form involves
a great deal of sophistication. With the exception of a few simple elements, the element
stiffness matrix for majority of elements is not available in explicit form. As such they
require numerical integration for their evaluation. The geometry of the element is defined
in reference to a global frame. In many problems such as those of rectangular plates, the
global and local axis systems are coincident and for them no further calculation is needed
at the element level beyond computation of element stiffness matrix in local coordinates.
Coordinates transformation must be done for all elements where in is needed.
 14

3.6.5. FORMULATION OF THE OVERALL STIFFNESS MATRIX

After the element stiffness matrices in the global coordinates are formed, they are
assembled to form the overall stiffness matrix. The assembly is done through the nodes,
which are common to adjacent elements. At the nodes, the continuity of the displacement
function and possibly their derivatives are established. The overall stiffness matrix is
symmetric and banded.

3.6.6. INCORPORATION OF BOUNDARY CONDITIONS

The boundary restraint conditions are to be imposed in the stiffness matrix. There are
various techniques available to satisfy the boundary conditions. In some of these
approaches, the size of the stiffness matrix may be reduced or condensed in its final form.
To ease the computer programming aspect and to elegantly incorporate the boundary
conditions, the size of the overall stiffness matrix is kept the same.

3.6.7. FORMULATION OF ELEMENT LOAD MATRIX

The loading forms an essential parameter in many structural engineering problems. The
loading inside the element is transferred at the nodal points and consistent element load
matrix is formed. Sometimes, based on the typicality of problem, the load matrix may be
simplified.

3.6.8. FORMATION OF THE OVERALL LOAD MATRIX

Like the overall stiffness matrix, the element loading matrices are assembled to form
the overall loading matrix. This matrix has one column per loading case and it is either
a column vector or a rectangular matrix depending on the number of loading conditions.

3.6.9. SOLUTION OF SIMULTANEOUS CONDITIONS

All the equations required for the solution of the problem are now developed. In the
displacement method, the unknowns are the nodal displacements. The gauss elimination
and Cholesky’s factorization are the most commonly used procedures for the solution of
simultaneous equations. These methods are well suited to a small or moderate number of
 15

equations. For large sized problems, a frontal technique is one of the methods of obtaining
solution. For systems of large order, Gauss-Seidel or Jacobi iterations are more suited.

3.6.10. CALCULATION OF STRESS OR STRESS-RESULTANTS

In the previous step, nodal displacements are calculated and these values are utilized for
the calculation of stresses or stress-resultants. This may be done for all elements of the
continuum or it may be limited only to some predetermined elements. Results may be
obtained by graphical means. It may be desirable to plot the contour of the deformed shape
of the continuum. The contour of the principal stresses may be one of the sought-after items
for certain category of problems.

3.7 ADVANTAGES

The main advantage of finite element analysis can be put in one sentence. The
physical problems which were so far intractable and complex for any closed bound solution
can now be analyzed by this method. The advantages in relation to the complexity of the
problem are stated below.

a. Finite element method is fast, reliable and accurate.

b. It can analyze any structure with complex loading and boundary conditions

i.e. the method can be efficiently applied to cater irregular geometry and can
handle any type of loading.

c. It can analyze structures with different material properties

i.e. material anisotropy and non-homogeneity can be catered without much

difficulty.

d. This method is easily amenable to computer programming.

e. It can analyze structures having variable thickness.
 16

3.8 DISADVANTAGES

One should not form the idea that the F.E.M is the most efficient for the analysis of
any type of structural engineering or physical problem. There are many types of problems
where some other method of analysis may prove efficient than the F.E.M.

1. Main disadvantages of this method are the cost involved in the solution of the
problem. (simpler computer methods such as finite strip or other same analytic
methods for the vibration and stability analysis of simpler structure will lead to
more economical solution, but these methods will work within their own
limitations and will not be as versatile as the F.E.M)
2. It is difficult to model all problems accurately and the results obtained are
approximate.
3. The result depends upon the number of elements used in the analysis.
4. Data preparation is tedious and time consuming.

3.9 GENERAL PROCEDURE

In the finite element method, the structure under consideration is divided into
smaller zones, called as elements. The elements are assumed are to be connected to each
other at certain points called nodes. It is at the nodes that we compute the displacements.
Thus, a body having degrees of freedom equal to two or three times the number of nodes
approximates the body with infinite number of degrees of freedom. With the increase in
number of nodes, a better solution (closer to the exact) is obtained. The displacements at
any point within an element are related to the displacements at the nodes by a set of
functions, the most common form being a polynomial. The polynomials chosen should be
such that they give continuity within the elements and the displacements must be
compatible between adjacent elements. The element stiffness matrices and the element load
vectors are formulated and assembled to get global vectors for the entire structure. After
applying the required boundary conditions, the formulated equations are solved to obtain
 17

the nodal displacements. From the displacement field within the element, strains can be
calculated. From the strains, using the stress- strain relations, stresses can be calculated.

START SCOPE

PRELIMINARY RESEARCH
• Identify problem
• Design structure

Create the Model

Analyze the Model Using Software

• Identify problem
• Design structure

Analyze the result, discussion and conclusion

Write thesis report COMPLETE

• Identify problem
• Design structure
Fig 3.1 METHODOLOGY
 18

CHAPTER IV

MATERIAL PROPERTIES

4.1 GENERAL

The set of wide-flange sections, which is investigated through detailed FE analysis,

comprises of 10 sections selected from the following sizes: ISHB200 to ISHB450. Based
on prior system level analytical studies that investigated the seismic behavior of steel SMFs
through, the selected sections reflect the typical members that are used as columns in low
to high-rise steel buildings with 2 to 6 stories designed in urban Chennai. In order to
investigate the effect of seismic compactness on the cyclic deterioration in flexural strength
of steel beam-columns, the remaining five sections are compact sections that meet the web
and flange width-to-thickness limits given for moderately ductile members, λmd, (AISC
2010a). These sections are used in intermediate moment frames. The flange and web
compact limits for highly ductile members are 7 and 43, respectively, assuming a modulus
of elasticity E = 207,980MPa (i.e., 29,000 ksi), expected yield stress Fy = 383MPa (i.e.,
Ry Fy = 55 ksi, in which Ry = 1.1) and axial load ratio P/Py = 0.5.

4.2 USE OF WIDE FLANGE STEEL BEAM AND COLUMNS

Wide flange columns with WB 150 to WB 600 sections provide larger moment of
inertia for the same weight compared to traditional WB 300 to WB 600 column sections.
For example, the weight of a WB 250 section will be less than weight of a WB 300 section
with comparable moment of inertia. Relatively large bending stiffness of the columns
results in increasing the global stiffness of the moment frame, which in turn results in
reducing the drift and damage.
 19

4.3 SECTION PROPERTIES TABLE

Table 4.1 Section Properties from IS 808: 1989

 20

Table 4.2 Loads on Column

Column Section Load on Structure (KN)

WB 150 10,667
WB 175 13,901
WB 200 14,617
WB 200 15,055
WB 225 16,098
WB 250 18,763
WB 300 20,244
WB 350 23,651
WB 400 26,912
WB 450 28,144
WB 500 30,405
WB 550 32,954
WB 600 36,256
WB 600 38,268
 21

CHAPTER V

FINITE ELEMENT MODELING AND VALIDATION

5.1 MODELLING APPROACH:

The main scope of this paper is to study the cyclic behavior of wide-flange steel
sections as part of first-story interior columns in typical SMFs designed in Chennai. To this
end, a FE model of a typical first-story interior column is developed in ABAQUS-
FEA/CAE (2018). The column has a length L = 4,600 mm (15 feet) as shown in Fig. 5.1.
This length represents typical first story heights in steel SMFs. Fully fixed boundary
conditions are applied at the base of the column (i.e., fixed support) while partially fixed
boundary conditions with flexible rotational stiffness are applied at the top of the column
(i.e., flexible support) in order to consider the flexibility of the beam-to-column connection
at the same location. These boundary conditions reflect more realistically first-story steel
columns in SMFs because they can capture the moment gradient changes along a first story
column once plastic hinging occurs at its base. Past analytical and/or experimental studies
either involved a cantilever or fixed-fixed boundary conditions.

Figure 5.1 Boundary conditions applied to the FE model

 22

For the fixed support, all six degrees of freedom are restrained. For the flexible
support, the out-of-plane rotation and displacement are restrained. The in-plane
displacement (Uy) is used to apply the lateral loading protocol. The in-plane rotation (Rx)
is controlled using a flexible elastic beam element (see Fig. 5.1). Depending on the flexural
stiffness of the elastic beam element we are able to control the partially rigid boundary
conditions due to the presence of the panel zone and beam-to-column connection that are
welded at both sides of an interior steel column as part of a steel SMF. For each column
section, the flexible beam element is assigned a pre-calculated in-plane moment of inertia
in order to keep the inflection point at a distance of 0.75 L measured from the base of the
column prior to the application of the lateral loading protocol. This is the expected location
of the inflection point in typical first-story columns when they remain elastic under lateral
loading. The flexible support boundary conditions are applied at a reference point (RP)
located at the center of the column top end (see Fig. 5.1). The column edge at the top end
is tied to the reference point using a “RIGID BODY” constraint in order to avoid possible
local stress concentrations.

Figure 5.2 Typical buckling mode used for local geometric imperfections
 23

The analytical model of the steel beam-column is meshed using a 25 mm × 25 mm

(i.e.,1”×1”) quadratic 4-node doubly-curved “S4R” shell elements. The S4R shell elements
can capture the expected severe local buckling within the cross section. These elements
also use reduced integration and hourglass control. To this end, they are computationally
efficient and they are able to avoid non-physical response modes. The 25 mm × 25 mm
mesh size is proposed after conducting a mesh sensitivity study not presented herein due
to brevity. In order to trigger local and global instabilities within the column section,
geometric imperfections are introduced to the FE model. These imperfections are
introduced to the FE model by scaling and super imposing different buckling modes.

Two types of imperfections are employed:

(1) local web and flange imperfections expected from the manufacturing process as
per limits (see Fig. 5.2);

(2) possible global out-of-straightness imperfections equal to L/1,[Link]

stresses are not considered.

in the modelling process because their effect is found to be negligible on the monotonic
and cyclic behavior of wide-flange steel columns.

Based on the boundary conditions of the proposed FE model, the location of the
inflection point changes once plastification occurs. Typically, once plastic hinging occurs
at the base of the column, the inflection point moves towards the base; hence the moment
demand at the top end of the column increases. For this reason, in order to calculate the
moment at the base, the location of the inflection point needs to be monitored throughout
the nonlinear analysis. To locate the inflection point, the normal stresses at equally spaced
cross sections along the column height are monitored as shown in Fig. 3b. For a 4,600 mm
long column, seven cross sections were found to be adequate for this purpose. The normal
stresses at each cross section are then used to compute the moment gradient along the
column height. Once the moment gradient is computed, the inflection point can be traced
as illustrated in Fig. 5.3a
 24

Figure 5.3 a) External forces acting on the deformed column, b) Monitoring of

moment gradient and inflection point location

with small scale slender cross sections, while the latter includes data from full-scale stocky
cross sections. In both experimental studies, various levels of axial load were applied to
the steel columns.
 25

CHAPTER VI
PARAMETRIC STUDY RESULTS AND DISCUSSION
6.1 CYCLIC DETERIORATION IN FLEXURAL STRENGTH OF WIDE-
FLANGE BEAM-COLUMNS
The end moment at both column supports is calculated as discussed in Sect. 5.1.
Figure 6.1 shows a typical example of the end moment versus the chord rotation (θ ) for a
WB 175 column when subjected to 20 % Py . The chord rotation, θ, is defined as the lateral
deformation at the top of the column, δ, divided with respect to the column length, L (see
Fig. 6.1).
The figure demonstrates how the FE model captures both the cyclic deterioration in
strength and stiffness of the column. The end moment-chord rotation hysteresis is used to
identify the flexural strength capacity (noted as Mmax) and the corresponding chord rotation
θmax prior to the occurrence of local buckling. For the sample hysteresis shown in Fig. 6.1,
the column reached a moment capacity =1.1 Mp when subjected to 20 % Py. Figure 6.1
also shows that the onset of local buckling occurs at the base of the column at a chord
rotation θmax = 1.5 % when subjected to 20 % Py . In Fig. 6.1, unloading stiffness
deterioration is also evident due to lateral torsional buckling of the steel column. Figure
6.2 shows the movement of the inflection point location that was monitored during the FE
analysis of the WB 175 column (see Fig. 5.3). Figure 6.2 shows that the inflection point is
located at 0.75 L at the beginning of the analysis (i.e., undamaged column). When
plastification occurs at the base of the column the inflection point moves towards the base
location. Note that by the end of the analysis, the inflection point stabilizes roughly about
the column mid-height. This is attributed to cyclic deterioration in strength at the base
location that triggers the initiation of plastification at the top location of the column as well.
Similar findings were observed for the other cases investigated as part of this paper.
 26

Figure. 6.1 End moment versus chord rotation for the W 175 column

Figure 6.2 Inflection point location versus chord rotation for the W 175 column
The end moment-chord rotation hysteresis in Fig. 6.1 is super imposed with a cyclic
envelope curve that connects the peak points of the hysteretic response of the steel beam-
column. The cyclic envelope curve is used to quantify the cyclic hardening and the cyclic
deterioration in flexural strength of a steel beam-column once local buckling occurs. It is
generally known that the envelope curve is loading history dependent.
however, it can still be used as an indicator to evaluate the steel column performance
under cyclic loading. Figure 6.3 (a to b) shows the cyclic envelope curves for each one of
the 40 cross sections that were analyzed as part of this paper (see Table 1) when subjected
to various axial load ratios. These curves are normalized with respect to the corresponding
expected plastic bending moment, Mp, for each one of the cross sections. Note that the
envelope curves are only shown in Fig. 7 for the chord rotation range between 1 and 6 %
radians. The effect of increasing compressive axial load (ranging from P/Py = 0 % to P/Py
= 50 %) on the cyclic envelope curves of the 40 sections is also shown in the same figure.
 27

Figure 6.3 Normalized peak flexural strength response versus chord rotation for the
load cases: a) 00% Py, b) 20% Py, c) 35% Py, d) 50% Py.
The main observations are summarized as follows:
• For cross sections with local slenderness ratios much smaller than the current
compactness criteria for highly ductile members (i.e., set W1 and similarly set F1),
cyclic hardening is about 60 % larger than that of cross sections that are near the seismic
compactness limits based on the same provisions (i.e., set W3 and similarly set F3).
This is attributed to: (1) the stocky web and flanges, of section sets W1 and F1, which
delay the onset of local buckling and hence flexural strength deterioration; (2) the large
number of small-drift inelastic cycles prior to the occurrence of local buckling based
on the symmetric lateral loading protocol.
 28

• In all cases, cyclic hardening decreases by about 40 % when the applied compressive
axial load increases from 0 to 50 % (see Fig. 6.3a, d). This is due to the early occurrence
of local buckling in the presence of high axial load ratios; hence strength deterioration
initiates at smaller chord rotations when the axial load ratio is increased. Cyclic
hardening is discussed in more detail in the following paragraph.
• The higher the axial load ratio the earlier the occurrence of local buckling. At 0 %Py
(see Fig. 6.3a), set W1 have an average θmax of 6.4 % radians compared to 1.3 % radians
when subjected to 50 % Py (see Fig. 6.3d).
• The smaller the local slenderness ratios of a cross section the larger the chord rotation
at which local instabilities are triggered. At P/Py = 50 % (see Fig. 6.3d), steel beam
columns that are part of set W1 have an average θmax of 1.5 % radians compared to 0.4
% radians for steel beam-columns part of set W3. Note that for P/Py = 50 %, compact
(λmd) cross sections (i.e., set W4 and similarly set F4) deteriorate rapidly and reach
zero flexural strength prior to 1 % radians; thus their envelop curves are not shown in
Fig. 6.3d.
• After the onset of local buckling, the more slender sections experience a more rapid
loss in flexural strength compared to the more compact sections. This can be observed
from the post-buckling slope of the envelope curves in Fig. 6.3 a–d. The cyclic
deterioration in flexural strength is assessed at the end of this section by quantifying the
chord rotation at which the column reaches 80 % Mp.
To further quantify the effect of cyclic hardening on the flexural strength of steel
beam columns, the ratio between the maximum moment, Mmax (as computed by FE
analysis), and the corresponding plastic flexural strength (Mp) is plotted in Fig 6.4a–d
versus the web local slenderness ratio of the 14 sections discussed in Table 4.1. Figure
6.4a–d, refers to axial loads equal to 0, 20, 35, and 50 % Py , respectively. Figure 6.4a
shows that at zero axial load (i.e., equivalent to a steel beam), highly compact sections (i.e.,
set W1) experience cyclic strain hardening up to 2Mp compared to 1.4Mp for the least
compact section sets W3 and F3. Figure 6.4d shows that at 50 % Py , the highly compact
section sets W1, and similarly F1, experience cyclic strain hardening levels of about 1.2
 29

while the moderately compact section sets W2 and F2 barely reached their plastic strength
capacity Mp. The least compact section set W3 and F3 and the compact (λmd ) W4 and F4
sets did not experience any cyclic strain hardening as they did not reach their expected
plastic flexural strength Mp in the presence of high axial load; this can also be seen in Fig.
6.3d.

Figure 6.4 Cyclic strain hardening versus web slenderness ratio for the load cases:
a) 00% Py, b) 20% Py, c) 35% Py, d) 50% Py.

Figure 6.5 Chord rotation at 80 % Mp versus web slenderness ratio for the load
cases: a) 00% Py, b) 20% Py, c) 35% Py, d) 50% Py.
 30

The cyclic hardening is an important contributor to member overstrength after

flexural yielding and may affect global collapse mechanisms in SMFs subjected to
earthquake loading. In particular, depending on the level of cyclic hardening, the
corresponding strong-column weak-beam ratio (SCWB) that a designer would use to select
a column at a particular interior frame joint may have to be increased. The reason is that
we tend to check the SCWB ratio based on the expected member plastic flexural capacities
(i.e., ignoring cyclic hardening);
However, in the case of beams (i.e., Fig. 6.4a) the corresponding beam flexural
capacities could be two times larger than Mp; in the case of columns, depending on the
level of the applied axial load ratio, the corresponding flexural capacities are not amplified
by much prior to the occurrence of local buckling. This necessitates a more careful revision
of the SCWB ratio in the future. Recent analytical studies on archetype steel buildings with
perimeter SMFs indicate that a more suitable SCWB ratio to be used for collapse mitigation
of steel SMFs is 2.0 instead of 1.0 if we also consider the effect of the composite action on
the cyclic behavior of steel beams. The previous findings have been also confirmed by a
coordinated analytical program related to the collapse behavior of full-scale steel frame
buildings tested to collapse.
However, columns are expected to undergo chord rotations larger than 3 % during
a maximum considered earthquake (i.e., 2 % probability of exceedance in 50 years) .Figure
6.5, suggests that currently employed seismic compactness requirements for highly ductile
wide flange sections per IS 808 : 1989 are satisfactory for upper stories in SMFs, in which
the axial load demands on the steel columns are smaller than 20 % Py . For interior columns
in bottom stories of SMFs, in which the level of axial load ratios ranges from 20 to 35 %,
a reduction to about two thirds of the current compactness limit for highly ductile wide-
flange sections would achieve a 4 % chord rotation with a reduction in column flexural
strength of no more than 20 % of the expected column plastic flexural strength, Mp. It
should be noted that the θ 80 %Mp values are dependent on the lateral loading protocol
(i.e., the SAC symmetric protocol). This indicates that a more thorough evaluation should
 31

be considered in which the effect of lateral loading protocol on the overall cyclic behavior
of a steel beam-column is taken into consideration.
6.2 PRE-CAPPING PLASTIC ROTATION
Structural engineers employ simplified numerical models in order to investigate the
seismic performance of new and existing steel frame buildings. These component models
need to be defined carefully in order to capture the nonlinear behavior of various structural
components as part of a steel building. These modelling options typically construct a multi-
linear “backbone” curve that bounds the force-deformation relation of a structural
component. As discussed earlier, experimental data is not available for deep slender wide-
flange columns undergoing combined lateral cyclic drift and axial load. The backbone
curve is also used to bound the cyclic behavior of a component model assigned to a steel
column as shown in Fig. 6.6a. The pre-capping plastic rotation (θp) is one of the main
parameters used to define the monotonic backbone curve along with other strength and
deformation parameters. The definition of the pre-capping rotation is shown in Fig. 6.6a.
These equations were obtained from extensive calibrations in which the parameters of the
backbone curve were matched to experimental data from steel beams with 20 < h/tw < 55
and 4 < b f /2t f < 8.

Figure 6.6 Definition of the pre-capping plastic rotation:

a) PEER/ATC, option 1, b) FEA monotonic loading case
 32

One of the main objectives of the FE study discussed in this paper is to evaluate the
analytical modelling options currently used by engineers to “realistically” model steel
columns. To this end, the pre-capping plastic rotation predicted by PEER/ATC (2010)
modelling option 1 (noted here as “θp, ATC-Option1”) is compared to the pre-capping
plastic rotation predicted by the FE analysis results. The latter is based on the monotonic
lateral loading case when combined with different axial load ratios (noted here as “θp,
MON-X”). The definition of the pre-capping plastic rotation obtained from the FE
monotonic curve is shown in Fig. 10b. Note that in some cases, especially at high axial
load ratios, the monotonic curve caps before reaching the expected plastic flexural strength
of the column (i.e., Mmax < Mp). In such cases the pre-capping plastic rotation is
considered to be equal to zero.
6.3 COLUMN PLASTIC HINGE LENGTH
The length of the “plastic hinge” formed at the base of a steel column during cyclic
loading is investigated in this section. The plastic hinge length is defined as the distance
from the base of the column to the cross sectional level with zero plastic strain. The
equivalent plastic strain (PEEQ) profile is monitored in 14 longitudinal sections along the
web and flanges of the set of 14 steel columns under investigation as shown in Fig. 6.7a.
This figure shows a comparison between the envelopes of the PEEQ strain profile as
distributed from the fixed end (i.e., location at 0 cm) till about 140 cm from the base
location for a WB 250 and a WB 200 steel column before reaching zero flexural strength,
when subjected to the load case 20%Py (i.e., P/Py = 20 %). Figure 6.7a shows that the least
compact WB 200 section (h/tw = 41.6 and b f /2t f = 4.44) develops a shorter plastic hinge
length compared to the WB 250 section (h/tw = 19.7 and b f /2t f = 3.19). The reason is
that due to early local buckling in the WB 250 section, plastic strains localize at a shorter
length and at smaller lateral drifts compared to the case of the WB 250 section. The plastic
hinge length of a stocky section spreads over a larger length after undergoing a larger
number of inelastic cycles compared to a more slender section. This is shown in Fig. 12b
in which a comparison between the progression of the plastic hinge length of the W30×391
and W30×148 columns, is plotted against the cumulative inelastic drift ( θH ) for the load
 33

case SYM-20 (i.e., P/Py = 20 %). The plastic hinge length of each cross section is
normalized with respect to the corresponding section depth. Note that at a given drift, θH
is calculated as the sum of absolute peak drift values of the inelastic cycles following the
yield drift rotation (θy ) of the respective column, where θy = Mp/Ke, in which Ke is the
initial elastic flexural stiffness of a column. By definition, the cumulative inelastic drift is
independent from the applied lateral loading protocol. Figure 6.6b shows that the
normalized plastic hinge length for the least compact section WB 200 stabilizes faster (at
θH = 0.35 radians) than the corresponding plastic hinge length of the more compact WB
225 section (at θH = 0.86 radians). Note that the effect of the compressive axial load ratio
is insignificant on the final stabilized plastic hinge length, given that that the column
underwent the sufficient number of inelastic cycles to reach the stabilized plastic hinge
length without losing its flexural strength.

Figure 6.7 Comparison between WB 225 and WB 200 beam-columns when subjected
to the load case 20% Py:
a) maximum PEEQ strain profile at the base of the column measured at zero
flexural strength
b) normalized plastic hinge length versus cumulative inelastic drift.
 34

Figure 6.8 Normalized plastic hinge length versus web slenderness ratio for the load
case 20% Py: a) measured at capping point Mmax , b) measured at zero flexural
strength
underwent the sufficient number of inelastic cycles to reach the stabilized plastic hinge
length without losing its flexural strength.
Figure 6.8a, b shows the plastic hinge length of all 12 sections shown in Table 4.1,
normalized by the corresponding column depth (d), versus the web local slenderness ratio
for the load case SYM-20 measured at Mmax (noted as L P H@Mmax ) and at zero flexural
strength (noted as L P H@Mzero), respectively. The L P H@Mmax values represent the
plastic hinge length before the onset of local buckling while L P H@Mzero values represent
the plastic hinge length after the localization and stabilization of local buckling. Figure 13a
confirms that highly compact sections develop a larger plastic hinge length compared to
slender sections. At 20 % Py , highly compact sections (i.e., h/tw < 22 and b f /2t f < 3.9)
develop an average plastic hinge length of about 2.0d compared to 1.3d for slender sections
(i.e., 32.5 < h/tw < 43 and 5.5 < b f /2t f < 7). The plastic hinge length is directly related to
the assumption of “point plastic hinge” models that have been widely used to assess the
cyclic behavior of steel columns as part of steel frame buildings. Based on the findings, for
highly compact members a distributed plasticity approach is more rational compared to
concentrated plasticity models.
 35

6.4 COLUMN AXIAL SHORTENING

The plastification of the column cross-section is coupled with the axial shortening
of the column member. The level of axial shortening becomes significant when severe local
buckling occurs, typically at high compressive axial loads. Consequently, axial shortening
could potentially be an important deterioration mechanism in SMF steel columns subjected
to extreme seismic events.
Figure 6.9a shows a comparison between the history of axial shortening versus the
chord rotation for the most and least compact WB 300 sections of Table 4.1, i.e., WB 300
(h/tw = 19.7 and b f /2t f = 3.19) and WB 250 (h/tw = 41.6 and b f /2t f = 4.44), respectively,
when subjected to the 20% Py load case. Axial shortening (axial) is defined as the axial
displacement of the column top end, measured at the reference point RP (see Fig. 5.1).
Note that in Figure 6.9a the WB 250 column reached zero flexural strength at 5 % radians
compared to 8 % radians for the WB 300. Figure 14a shows that at the second cycle of 4
% radians, the WB 300 column experiences axial shortening equal to 73 mm (1.6 % L)
compared to 166 mm (3.6 % L) for the WB 200 section. Axial shortening increases
substantially when the column is undergoing more inelastic cycles after the occurrence of
flange and web local buckling.

Figure 6.9 a) Comparison of axial shortening versus chord rotation for WB 300 and
WB 200 columns when subjected to 20% Py
b) Local buckling at the base of the WB 300 and WB 200 columns at 4 % drift when
subjected to 20% Py
 36

Figure 6.10 Normalized axial shortening versus cumulative inelastic drift for the
load cases: a) 0% Py, b) 20% Py, c) 35% Py, d) 50%Py.
This explains the larger amount of axial shortening of WB 200 column where local
buckling occurred at 1.4 % radians compared to 3.0 % for the WB 300 column, for the 20%
Py load case. The severe local buckling occurring in the slender WB 200 column is shown
in Figure 6.8b in comparison with that of the WB 300column at the end of the second 4 %
drift cycle. It is also worth mentioning that for the 20%Py load case, the most compact
section WB 175 (h/tw = 12.1 and b f /2t f = 2.15) experiences an axial shortening equal to
1 % L compared to 6.5 % L for the least compact section W36×230 (h/tw = 41.4 and b f
/2t f = 6.54).
Figure 6.9 shows the normalized axial shortening ratio (axial/L), versus the
cumulative inelastic drift ( θH ) for the applied load cases (i.e., P/Py = 0, 0.20, 0.35 and
0.50). A vertical line at θH = 0.7 is also superimposed in Fig. 15. This represents the
average cumulative inelastic drift value, for all sections, that corresponds to the second
cycle of magnitude 4 % radians. The θH = 0.7 is used as a reference to quantify the level
 37

of axial shortening expected in first-story columns at 4 % chord rotations. At 0 % Py and

θH = 0.7 (see Fig. 6.10a), low levels of axial shortening ratios (less than 1 %) are observed
in all compact sections. This is attributed to the fact that the majority of compact sections
reached their cyclic maximum flexural strength at chord-rotations larger than 4 %. Once
the axial load is applied to the columns, the corresponding cross section plastifies at fairly
low plastic rotations compared to the case with P/Py = 0.0. The occurrence of local
buckling leads to significant amplification of column axial shortening. At 20 % Py , sets
W1 and F1, (i.e., h/tw < 22 and b f /2t f < 3.9) experience an average axial shortening ratio
equal to 2.5 % compared to 6 % for section set W3 and F3 (i.e., 32.5 < h/tw < 43 and 5.0
< b f /2t f < 7.0). The compact (λmd ) sections (i.e., set W4 and set F4) failed before
reaching the second 4 % cycle (see Fig. 6.10b). It should be noted that the variation in axial
shortening ratio becomes considerable at higher drifts and higher axial loads even between
sections within the same set. This is attributed to the exponential increase in axial
shortening, once local buckling occurs, with cumulative inelastic drift. This is shown in
Fig. 6.10b–d. Note that set W1 undergoes larger number of inelastic cycles and hence
experiences higher level of axial shortening before reaching zero flexural strength,
compared to W3 (see Fig. 6.10b–d).
 38

CHAPTER VII
CONCLUSION

This paper investigates the cyclic behavior of wide-flange first-story interior steel
columns as part of steel special moment frames (SMFs) designed in highly seismic regions.
First, a detailed finite element (FE) modelling approach is proposed in order to model steel
beam columns. The modelling approach is validated with past experimental cyclic test data.
The FE model was able to successfully reproduce the experimental cyclic moment-rotation,
axial shortening and longitudinal flange strain for all the specimens from two different
experimental programs. The same FE modelling approach is used to model a typical first-
story interior steel column, in which the flexibility of the column top-joint is taken into
account. The column FE model is used to investigate a set of 40 wide-flange sections
covering a wide spectrum of web and flange slenderness ratios. The columns are subjected
to symmetric cyclic lateral loading combined with compressive axial load levels ranging
from 0 to 50 % Py . Deterioration mechanisms that are directly related to dynamic collapse
of steel SMFs are investigated, such as cyclic deterioration in flexural strength and column
axial shortening. The plastic hinge length developed at the column base is also quantified.
In addition, the pre-capping plastic rotation capacity is quantified. The major findings are
summarized as follows:

• In the absence of axial load, compact wide-flange sections achieve significant

levels of cyclic strain hardening. For highly compact sections (i.e., h/tw < 22 and
b f /2t f < 3.9), the maximum moment Mmax can reach up to two times the
expected plastic flexural strength, Mp, of a steel cross section compared to
1.3Mp for sections with large local slenderness ratios (i.e., 32.5 < h/tw < 43 and
5.5 < b f /2t f < 7).
• In the presence of axial load, the level of cyclic stain hardening decreases with
the increase of compressive axial load ratios. At an axial load level of 50 % Py ,
highly compact sections reach an average cyclic strain hardening of 1.2Mp while
 39

low or moderately compact sections (i.e., 22 < h/tw < 43 and 3.9 < b f /2t f < 7)
were not able to reach their expected plastic flexural strength (i.e., Mmax < Mp).
These findings have direct implications on the strong-column-weak-beam
(SCWB) ratio used in the seismic design of steel SMFs; normally, we tend to
ignore cyclic hardening of the corresponding steel members in the SCWB check.
• Currently employed seismic compactness requirements for highly ductile wide-
flange sections per IS 808 1989 are satisfactory for interior columns in upper
stories of SMFs, in which the axial load demands on the columns are smaller
than 20 % Py . For interior steel columns in bottom stories of SMFs, in which
the level of axial load ratios are in the range of 20–35 % Py , a reduction to about
two thirds of the current compactness limit for highly ductile wide-flange
sections would achieve a 4 % chord rotation with a reduction in column flexural
strength of no more than 20 % Mp.
• In the absence of axial load, the pre-capping plastic rotation predicted by
PEER/ATC (2010), modelling option1, seems to be a reasonable estimate for
cross sections with 32.5 < h/tw < 43 and 5.5 < b f /2t f < 7 undergoing monotonic
loading and underestimated for moderate and highly compact sections (i.e., h/tw
< 32.5 and b f /2t f < 5.5).
• PEER/ATC (2010), modelling option1, reasonably estimates the pre-capping
plastic rotation for all compact cross sections with h/tw < 43 and b f /2t f < 7
undergoing cyclic loading and axial load levels up to 35 % Py . The same
recommendations seem to overestimate the pre-capping plastic rotation of steel
components with local slenderness ratios ranging from 32.5 < h/tw < 43 and 5.5
< b f /2t f < 7 subjected to either monotonic or cyclic lateral loading combined
compressive axial load ratios in the range of 35–50 % Py .
• Highly compact steel columns (i.e., h/tw < 22 and b f /2t f < 3.9) develop an
average plastic hinge length of 2.0d compared to 1.3d for slender sections (i.e.,
32.5 < h/tw < 43 and 5.5 < b f /2t f < 7) regardless of the applied axial load ratio.
 40

The buckling length of the same columns is in average equal to the depth, d, of
the corresponding cross section.
• At 20 % Py , columns with highly compact cross sections (i.e., h/tw < 22 and b
f /2t f < 3.9) experience an average axial shortening ratio equal to 2.5 % of their
length compared to about 6 to 8 % of highly ductile wide-flange sections per
AISC-341-10 (i.e., 32.5 < h/tw < 43 and 5.5 < b f /2t f < 7).

It should be noted that the results presented here are dependent on the applied loading
protocols and the selected beam-column boundary conditions. The authors currently
evaluate the behavior of the same set of beam-columns for other lateral and axial loading
protocols. The findings from the extensive analytical study summarized herein will also be
complemented by a series of full-scale experiments on deep wide-flange beam-columns.
 41

CHAPTER VIII

REFERENCE

1) Dafalias Y, Popov E (1975) A model of nonlinearly hardening materials for

complex loading. J Acta Mech 21(3):173–192. doi:10.1007/BF01181053
2) de Souza Neto EA, Peri´c D, Dutko M, Owen DRJ (1996) Design of simple low
order finite elements for large strain analysis of nearly incompressible solids. Int J
Solids Struct 33(20–22):3277–3296. doi:10.1016/ 0020-7683(95)00259-6
3) Elkady A, LignosDG(2012) Dynamic stability of deep slender steel columns as part
of special MRFs designed in seismic regions: finite element modeling. In: Proc. first
international conference on performance-based and life-cycle structural engineering
(PLSE), Hong Kong, December 5–7, 2012
4) Elkady A, Lignos DG (2014) Modeling of the composite action in fully restrained
beam-to-column connections: implications in the seismic design and collapse
capacity of steel special moment frames. J Earthq Eng Struct Dyn.
doi:10.1002/eqe.2430
5) FEMA (2000) State of the art report on connection performance, Report FEMA-
355D. Federal Emergency Management Agency, Washington
6) Gupta A, Krawinkler H (1999) Seismic demands for the performance evaluation of
steel moment resisting frame structures, Report No. 132. The John A. Blume
Earthquake Engineering Center, Stanford University, Stanford
7) IS 800:2007, Indian Standard Code Of Practice For Construction.
8) is 808:1989, Indian Standard Dimensions For Hot Rolled Steel Beam, Column,
Channel And Angle Sections
9) Kaufmann EJ,Metrovich BR, Pense AW(2001) Characterization of cyclic inelastic
strain behavior on properties of A572 Gr. 50 and A913 Gr. 50 rolled sections,
ATLSS Report 01–13. National Center for Engineering Research on Advanced
Technology for Large Structural Systems, Bethlehem
 42

10) Krawinkler H, Zohrei M, Lashkari-Irvani B, Cofie NG, Hadidi-Tamjed H (1983)

Recommendations for experimental studies on the seismic behavior of steel
components and materials, Report No. 61. The John A. Blume Earthquake
Engineering Center, Stanford University, Stanford
11) Lignos DG, Krawinkler H (2011) Deterioration modeling of steel components in
support of collapse prediction of steel moment frames under earthquake loading. J
Struct Eng 137(11):1291–1302. doi:10.1061/ (ASCE)ST.1943-541X.0000376
12) Lignos DG, Hikino T, Matsuoka Y, Nakashima M (2013) Collapse assessment of
steel moment frames based on E-defense full-scale shake table collapse tests. J
Struct Eng 139(1):120–132. doi:10.1061/(ASCE)ST. 1943-541X.0000608
13) Lignos DG, Krawinkler H (2013) Development and utilization of structural
component databases for performance-based earthquake engineering. J Struct Eng
139(8):1382–1394. doi:10.1061/(ASCE)ST. 1943-541X.0000646
14) MacRae GA, Hyde K,WalpoleWR, Moss PJ, Hyland C, Clifton GC, Mago N (2006)
Column axial shortening effects in steel frames. In: Proceeidngs of New Zealand
society for earthquake eng. conference, Napier, New Zealand, March 10–12, 2006
15) PEER/ATC (2010) Modeling and acceptance criteria for seismic design and
analysis of tall buildings, PEER/ATC 72–1. Prepared for Pacific Earthquake
Engineering Research Center (PEER) by Applied Technology Council (ATC),
Redwood City, San Diego
16) Steel Table.
17) Zareian F, Lignos DG, Krawinkler H (2010) Evaluation of seismic collapse
performance of steel special moment resisting frames using FEMA P695 (ATC-63)
methodology. Proc. ASCE Structures Congress, Orlando, May 12–15, pp 1275–
1286. doi:10.1061/41130(369)116
18) Zhang X, Ricles JM (2006) Experimental evaluation of reduced beam section
connections to deep columns. J Struct Eng 132(3):346–357.
doi:10.1061/(ASCE)0733-9445(2006)132:3(346)