Scribd
Upload
27 views12 pages

MSA Lecture 4 240807

Uploaded by

vikramsinghmeena414
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
27 views12 pages

MSA Lecture 4 240807

Uploaded by

vikramsinghmeena414
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 12

MATRIX STRUCTURAL

ANALYSIS (CE32201)
ARITRA CHATTERJEE
DEPARTMENT OF CIVIL ENGINEERING
AUTUMN 2024
LECTURE 4 – GENERAL ALGORITHM FOR MATRIX STRUCTURAL ANALYSIS
(FOR 2D TRUSSES)
STEP 1 – REPRESENT THE STRUCTURE
WITH MEMBERS AND NODES
u Line diagram based representation
u All nodes and members are to be numbered
u Boundary conditions to be drawn
u Good practice to follow a logical flow for numbering, for example, cyclical
and counter-clockwise

Reference : Matrix Analysis of


Structures by Aslam Kassimali
STEP 2 – CREATE A NODAL COORDINATES
MATRIX
u Against each node number, record its x, y (z if 3D) coordinates
u For 2D problems, this could be a N*3 matrix. First column is node number,
second column is x coordinate, third column is y coordinate
u N rows represent the N nodes in the problem
STEP 3 – CREATE A MEMBER
CONNECTIVITY AND PROPERTY MATRIX
u Against each member number, record its starting node (or i node) and ending
node (or j node)
u Also record the material and geometrical properties of each member. For
truss problem, these would be modulus of inertia E and cross-sectional area A
u This could be a M*7 matrix for a truss problem. First column is member
number, second column is ith node, third column is jth node, fourth column is E
and fifth column is A
u Sixth column is member length and seventh column is member angle theta.
These will be calculated and not assigned (as discussed in the next slide)
u M rows represent M members in this problem
STEP 3 (CONTD.) – CALCULATE AND SAVE
MEMBER LENGTH AND ANGLE
u Member length is calculated by finding the distance between coordinate of
node i and node j
u Angle theta can be calculated using tan-1[(yj-yi)/(xj-xi)]
u NOTE 1- Be careful with radian and degree, depending on the program used
u NOTE 2 – Do we need to track theta < pi radians or theta > pi radians? Why or why
not?
u These quantities need to calculated by referring to the ith and jth nodes of
nodal coordinates matrix!
STEP 4 – ASSEMBLE THE GLOBAL
UNRESTRAINED STIFFNESS MATRIX
u First initialize the full matrix with zeros
u Size of matrix is 2*N X 2*N (WHY???)
u For each member (that is, each row of member connectivity and property
matrix):
u Generate the 4X4 truss member stiffness matrix in local coordinate system
u Using the member angle theta, generate the coordinate transformation matrix [T]
u Calculate the 4X4 member stiffness matrix in global coordinate system by using
[T]T * [k4X4] * [T]
u Map the local DOFs to global DOFs (1 is mapped to 2i-1, 2 is mapped to 2i, 3 is
mapped to 2j-1, 4 is mapped to 2j)
u Based on the map, add the terms of the 4X4 member stiffness matrix to the global
unrestrained stiffness matrix
u (Repeat for all members)
STEP 5 – GENERATE THE GLOBAL
RESTRAINED STIFFNESS MATRIX
u Identify the degrees of freedom which are restrained
u Delete the corresponding rows and columns from the unrestrained stiffness
matrix and create the restrained stiffness matrix
STEP 7 – CREATE THE FORCE VECTOR

u Create a column vector with applied force at each unrestrained DOF


u Delete each DOF from the column vector where a kinematic boundary
condition is applied
u Finally, number of columns should match of force vector should equal the size
of the unrestrained stiffness matrix
STEP 8 – INVERT THE GLOBAL
UNRESTRAINED STIFFNESS MATRIX
u For practice, you may use an in-built function in the program of your choice
u For learning (and assignment / exam etc.), you need to write your Gaussian
elimination program as a function, then pass the unrestrained stiffness matrix
to the function as an argument, and recover the inverted matrix
STEP 9 – CALCULATE UNKNOWN
DISPLACEMENTS
u Multiply inverted stiffness matrix with global force matrix to get
displacement vector
u Plot displaced shape of structure
STEP 10 – FIND INTERNAL MEMBER
FORCES
u Map the displacement vector to the DOFs of the member desired
u Transform displacements from global to local coordinates using [T] * {v}
u Multiply displacements in local coordinate system with local stiffness matrix
to get member forces
u Forces will represent tension and compression based on your connectivity
definitions (local DOF definitions)
STEP 11 – FIND SUPPORT REACTIONS

u Multiply global unrestrained with full displacement vector (including unknown


displacements, and zeros at restraints)
u Read forces at restrained DOFs – these are the support reactions

You might also like