Analysis of Structures by the Finite Element Method
Index:
Element Flexibility Matrix and Element Stiffness Matrix
Analysis of Statically Determinate Structures by the Force Method
Analysis of Statically Indeterminate Structures by the Force Method
Notion of the Primary Structure
Analysis of Structures by the Displacement Method
Structure Flexibility Matrix and Stiffness Matrix
Comparison of Force and Displacement Methods
The structure is idealized as the composition of a number of finite elements rather than as a continuum of pieces.
The structure’s external forces (or loads) are represented by a column matrix (or vector) R for each node 1,2, … etc..

The structure’s corresponding nodal displacements are represented by a column matrix r for each node 1, 2, … etc.

The structure’s internal end forces are represented by a column matrix Q for each members a, b, … etc.

The structure’s corresponding internal nodal displacements are represented by a column matrix q for members a, b, … etc.
where the superscript a, b, c, … etc. denote the element while the subscript i denotes the near-end node and i and the far-end node of the element.
For each element, the internal nodal displacements caused by external forces (loads) are threefold: (1) rotation at the near-end node i, (2) rotation at the far-end node j, and (3) elongation k (shortening) of the element.

In matrix notation.
![]()
in which
is defined as the element flexibility
matrix.

For a structure consisting of a, b, … etc. elements

In matrix form
![]()
where f is a diagonal matrix with element flexibility matrices as diagonal constituents.
Alternatively, for each element a, b, … etc. the internal forces caused by nodal displacements are threefold: (1) moment at the near-end node I, (2) moment at the far-end node j and (3) tension (compression) in the element itself.
.
In matrix notation
![]()
in which
is defined as the
element stiffness matrix.

For a structure consisting of a, b, … etc. elements we have

In matrix form
![]()
where k is a diagonal matrix with elemental stiffness matrices as it constituents.

Element Flexibility Matrix and Stiffness Matrix
It can be readily shown that he element flexibility matrix is the inverse of the element stiffness matrix.
![]()
The principle of virtual work states that for an elastic structure in equilibrium, the external virtual work done in deforming the structure must be equal to the internal virtual work done by the structure in resisting deformation.
![]()
Or by taking the transform of both sides of the above equation
![]()
Analysis of Statically Determinate Structures by the Force Method
For statically determinate structures, the internal forces
( i= 1, 2, … , m) can be expressed in terms of the
external nodal forces (loads)
(
j= 1, 2, …, n) by using equilibrium conditions alone

In matrix form
![]()
where b is the structure force transformation matrix (m, n)

The structure nodal displacements q are related to the member internal forces Q through the flexibility matrix f
![]()
Substituting for Q in above eqution
![]()
Using the principle of virtual work
![]()
Substituting for Q and q in the right hand side of the above equation
![]()
![]()
![]()
The above equation solves for nodal displacements r in terms of nodal external forces R
![]()
where F is defined as the total flexibility matrix of structure
![]()
Summary:
Define the external nodal forces (loads) R
Define the internal member forces Q
Determine the force transformation matrix b. Equilibrium and the
superimposition principle can be used to determine the elements of b: The
external forces (loads) R, if taken one at a time, will determine the member
internal forces Q by using the relationship
to give
the elements of b.
Determine the individual flexibility matrices
and assemble them into the diagonal matrix

Compute the structure flexibility matrix
![]()
Find the nodal displacements ![]()
Analysis of Statically Indeterminate Structures by the Force Method
For statically indeterminate structures, the external virtual work done in deforming the structure must include the work done by the forces in the redundant elements as well
![]()
For equilibrium, the external and internal forces must balance
![]()
Taking the transpose of both sides of the above equation
![]()
Recall that ![]()
Substituting for Q in the right side of the above equation
![]()
Substituting these equations into the equation of virtual work
![]()
Expanding the right side of the above equation and comparing virtual forces on the left side and right side of the equation, we have

In matrix form

For structures on rigid supports
![]()
Therefore
![]()
![]()
The above equation is the expression for the redundant forces X.
The nodal displacements are given by
![]()
or ![]()
where
is the flexibility matrix of the redundant
structure
With the redundant forces X found, the member forces are solved by equilibrium
![]()
Substituting for X in the above equation
![]()
or ![]()
where
is the force transformation matrix of the
indeterminate structure.
Summary:
Define the external loads R
Define the internal member forces Q, and specify the redundant forces X
Calculate the force transformation matrices
and
from equilibrium
![]()
Determine the individual flexibility matrices
and assemble them into the diagonal matrix

Calculate ![]()
Calculate ![]()
Find the inverse of ![]()
Solve for the redundant forces ![]()
Substitute X into the equilibrium equation to find the member forces
![]()
Calculate ![]()
Find the nodal displacements
![]()
Notion of the Primary Structure
In the analysis of statically indeterminate structures by the Force Method, recall that
![]()
This says that the internal forces Q are linearly related to a set of applied external forces R and a set of unknown forces X associated with the redundant members of the structure.
If the redundant elements are removed one by one, the structure remains stable up to the point at which the last remaining redundant element is removed. What remains once the redundant elements are removes is the notion of the primary structure.
Structures in practice are usually highly redundant in their configurations. Redundant elements (bars) add strength and stability to the structure usually at an increment to the overall cost of the structure.
Analysis of Structures by the Displacement Method
For a structure, the member deformations q should be consistently related to the nodal displacements r such that the elements of the structure fit together at the nodal points

In matrix form
![]()
where a is the structure displacement transformation matrix a (m, n).

From the principle of virtual work
![]()
Recall
and taking the transform of both sides
![]()
Substituting for Q and q in the right hand side of the equation of virtual work
![]()
from which ![]()
Recall
. Taking the transform of both sides
![]()
Recall
and ![]()
![]()
Substituting for Q and q in the equation of virtual work
![]()
![]()
![]()
where K is defined as the total stiffness matrix of the structure
![]()
In the approach known as the Direct Stiffness Method, K is assembled
progressively from scratch by substituting and adding the normalized stiffness
matrices
(i=1,2, …) of each element in the correct
matrix locations of the structure total stiffness matrix K.
Solve for displacements r using the external forces (loads) R using
![]()
Solve for the internal member forces Q using
![]()
Summary:
Define all possible nodal displacements r. For a 3D truss, a pin connected node has three displacements: three linear (in the x, y and z directions of the members local coordinate system). For a 3D frame, a rigidly connected node has six displacements: three linear and three rotational (about the x, y and z axes of the member’s local coordinate system).
Determine the displacement transformation matrix a from geometric considerations
From the individual stiffness matrices
obtain

Compute the total stiffness matrix of the structure
![]()
Obtain the inverse of K
Compute the nodal displacements by
![]()
Compute the member forces by ![]()
Structure Flexibility Matrix and Stiffness Matrix
It can be readily shown, that he structure stiffness matrix K is the inverse of the structure stiffness matrix F
![]()
Comparison of the Force Method and the Displacement Method
| Force Method | Displacement Method | |
| 1 | Select member forces as basic unknowns | Select nodal displacements as basic unknowns |
| 2 | Establish the force transformation matrix for each element | Establish the displacement transformation matrix for each element |
| 3 | Evaluate member flexibility matrices for each element | Evaluate member stiffness matrices for each element. |
| 4 | Obtain the structure total flexibility matrix | Obtain the structure total stiffness matrix |
| 5 | Express nodal displacements in terms of nodal forces | Express nodal forces in terms of nodal displacements |
In the displacement method, the irrelevancy of redundancy enables the use of the same procedures for analyzing statically-determinate and statically-indeterminate structures. This fact taken together with the Direct Stiffness Method of assembling K gives the Displacement Method some real advantages over the Displacement Method.
Last updated: 02/12/2007