Scribd
Upload
748 views

Presentation On Matrix Method For Structure Analysis

The document discusses two research papers on analyzing non-prismatic structural members. The first paper presents a flexibility-based method using basic displacement functions to analyze geometrically nonlinear non-prismatic beams. The second paper derives general expressions for the stiffness matrix and load vector of continuous and discontinuous non-prismatic members using a transfer matrix method. It shows this method is more flexible and applicable than other procedures for stiffness matrix derivation. The document also notes that further research could include comparing the transfer matrix method results to software solutions and testing its effectiveness on other structural components.

Uploaded by

VAIDEHI BEAUTY CARE by DEEPA BHAVSAR
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
748 views

Presentation On Matrix Method For Structure Analysis

The document discusses two research papers on analyzing non-prismatic structural members. The first paper presents a flexibility-based method using basic displacement functions to analyze geometrically nonlinear non-prismatic beams. The second paper derives general expressions for the stiffness matrix and load vector of continuous and discontinuous non-prismatic members using a transfer matrix method. It shows this method is more flexible and applicable than other procedures for stiffness matrix derivation. The document also notes that further research could include comparing the transfer matrix method results to software solutions and testing its effectiveness on other structural components.

Uploaded by

VAIDEHI BEAUTY CARE by DEEPA BHAVSAR
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 20

PRESENTATION ON MATRIX

METHOD FOR STRUCTURE


ANALYSIS
NISARG D. BHAVSAR
FACULTY ADVISOR: Prof. SATYEN RAMANI
DEPARTMENT OF STRUCTURE ENGINEERING
SAL INSTITUTE OF TECHNOLOGY AND ENGINEERING RESEARCH
RESEARCH PAPER I
Geometrical Nonlinear Analysis of Non-
prismatic Beams Using Basic Displacement
Functions

S.M. Bagheri and R. Attarnejad


School of Civil Engineering, University College of Engineering, University of Tehran, Iran.
INTRODUCTION

 The geometrical non-linear analysis of the frames is an elastically instability


problem. In spite of the fact that the structure is still elastic, the large
deflection effects cause to transfigure of the structure, hence the linear
elastic theory becomes invalid.
 Most of the literatures on non-prismatic beams have displacement-based
formulations established on an assumed or prescribed displacement field.
This makes an extra hypothesis usually imposed in addition to the three
essential relations: equilibrium of forces, compatibility of displacements
and/or strains and the constitutional law of the material behaviour.
STRUCTURAL FORMULATION
(Basic Displacement Functions)
 BDFs are mathematical functions with mechanical concepts. These
functions represent displacement of the free end of a cantilever beam due
to a unit load at distance x.
 BDFs can be calculated from the relations shown in table I.
STRUCTURAL FORMULATION
(SHAPE FUNCTION)
 As described, the nodal Softness matrices of the element can be obtained
using the BDFs.

 Shape functions derive out from the following relation:


 U(x)=[Nu]{d}={Bu} ͭ [G]{d}
 W(x)=[Nu]{d}={Bf} ͭ [G]{d}
STRUCTURAL FORMULATION
(GEOMETRICAL STIFFNESS MATRICES)
 The principle of stationary potential energy provides a
convenient way for developing a force- deflection
relationship for a structural element.
 Applications of principle of stationary potential energy lead
to the expression
 [k₁]{Δd}={Δf}
 Where [k₁] is the incremental stiffness matrix that relates the
incremental displacement {Δd} to the corresponding
incremental element forces{Δf}. This matrix is made up of four
distinct matrices.
 [k¡]=[k₀]+[kp]+[k₁]+[k₂]
STRUCTURAL FORMULATION
(GEOMETRICAL STIFFNESS MATRICES)
 In which [k₀]is the conventional linear stiffness matrix for uncoupled bending and
axial behavior. Its terms are constant; [kp] is linear function of axial force present
at the beginning of the incremental step, and [k₁]and [k₂]consist of terms that are
linear and quadratic functions of the incremental element displacements.
 Using BDFs, mentioned matrices are
calculated from equations below
SOLUTION PROCEDURE

 The “direct method” has been chosen in this paper to analyze the
geometrical nonlinear behavior of a structure.
 [ks]=[k₀]+[k₁]+[k₂]
 Since [k₁] and [k₂] are function of unknown displacement, a straight
iteration procedure is used to solve above equation for the displacement
[Df] corresponding to a given load[Fa].
VALIDATION TEST

 To check the competency of the present method, the Wiliam’s frame has
been used as a validation test. The result of the presented method in this
paper compared by the result of common finite element procedures and
the analytical solution by wiliam’s frame.
SUMMARY OF STRENGTH

 The presented flexibility- based method and displacement-based methods


provide similar results in constant area section problems. But in non-
prismatic beams, the flexibility- based method results much better; due to
the exact shape functions derived using BDFs. The significant property of
these shape functions is their susceptibility to the variations in cross sectional
area, moment of inertia and even modulus of elasticity along the axis of
the beam element.
RESEARCH PAPER II
Accurate Stiffness Matrix for Nonprismatic
Members

Yao-Zhi Luo, Xian Xu and Fang Wu


ISSN 0733-9445
INTRODUCTION

 Generally speaking, there are two types of nonprismatic members, namely


continuous members and discontinuous members. A discontinuous
member usually consists of some continuous sub-members. The aim of the
note is to deduce general expressions of the stiffness matrix and equivalent
node load vector for both continuous and discontinuous nonprismatic
members with transfer matrix method (TMM) .
 Compared to other procedures for stiffness matrix derivation, TMM is more
flexible, and more applicable for computer programming.
METHODOLOGY
PRINCIPAL OF TRANSFER MATRIX METHOD

 The coordinate system used in the paper is defined in Fig. 1, together with
the definition for the positive directions of nodal forces and displacements.
The state vector which is a combination of the force vector
F={NxQyQzMxMyMz }and displacement vector = {uvwuxuyuz }is given by
 S = {Nx Qy Qz Mx My Mz u v w ux uy uz }T
TRANSFER OF STATE FACTOR
 The strain-displacement relation for a three-dimensional 3D member is
given by

 Where, γxy and γxz = shear strains in the x-y plane and x-z plane,
respectively; kx = rate of the change of twist; and ky and kz = curvatures in
the x-y plane and x-z plane, respectively.
 The force-strain relation for a 3D member is given by

 where EA, GA /μ , and GJ = axial stiffness, the shear stiffness, and the
torsional stiffness
 Consider a beam with its centroidal axis coinciding with the x axis of the
local coordinate system and with its left end located at the origin.

 Where,

 f = {qx qy qz mx my mz 0 0 0 0 0 0 }T
 The same positive directions are used for the distributed forces & moments
solving above equation we get

 The transfer matrix T x and T x-t are given by


Derivation of Transfer Matrix
Transfer Matrix for Continuous Members
 For continuous members, whose cross-sectional variables are continuously
changing along the lengths, the transfer matrix can be obtained by directly
solving the differential equation of state vectors, as presented in the last
section.
 Besides the above-mentioned procedure, the transfer matrix also can be
derived based on the principle of equilibrium and the principle of virtual
work, which will be presented next.
 When f= 0, the transfer matrix represents the transfer relation-ship between
the state vectors for the left node and right node.

 where the transfer coefficient ti,j represents the transfer relation-ship


between the state variables of the left node slj and the state variables of
the right node sri.
Derivation of Transfer Matrix
Transfer Matrix for Discontinuous Members
 A discontinuous member can usually be treated as a combination of
several continuous segments. And for each continuous segment, its transfer
matrix can be obtained by the same means as used for continuous
members. Then the transfer matrix for the entire member can be
determined from the transfer matrices of the segments.
 It is assumed that the two segments are subjected to span loads f1 and f2,
respectively. As shown in figure, the beam can be divided into two
substructures.

 Where,T1 and T2 = transfer state vectors of the two segments, respectively,


and Sq1 and Sq2 = transferstate vectors of the two segments, respectively

 We get where t1, Sq1 are,


SUMMARY OF STRENGTH
 The proposed method is based on the fact that the stiffness matrix and
equivalent node load vector can be written in the form of the transfer
matrix and transfer state vector.
 Compared with other methods which are only applicable to the stiffness
matrix derivation of continuous members, the TMM is a general method
that is applicable to the derivation of the stiffness matrix and the equivalent
node load vector for both continuous and discontinuous nonprismatic
members.
 For continuous members, the transfer matrix can be directly obtained by
solving the differential equation of state vectors or using the principle of
equilibrium and the principle of virtual work.
 For discontinuous members, the transfer matrix can be obtained by
accumulatively combining the transfer matrices of the continuous
segments. Combining the derived exact expressions with numerical
methods such as FEM will improve the precision and efficiency of the
numerical methods.
RESEARCH GAP

 Here the author has compared the transfer matrix method with virtual work
method, but has not compared to the software result. With the help of the
software result we can compare results precisely.
FURTHER RESEARCH

 For further research on this topic we can compare result of transfer matrix
method to software result.
 Furthermore, we can check that transfer matrix method is effective for
other structural member.

You might also like