Chapter Three: Matrix Methods for Beam Analysis

CHAPTER THREE:

Matrix Methods for Beam Analysis

In this chapter, we will explain the basic fundamentals of using the Matrix method or
stiffness method for analyzing structures. The method will be applied to different types of
Beams.

1. Preliminary Remarks

1.1.Member and Node Identification

In order to apply the stiffness method to beams, we must first determine how to
subdivide the beam into its component finite elements. In general, each element must be free
from load and have a prismatic cross section. For this reason the nodes of each element are
located at a support or at points where members are connected together, where an external force
is applied, where the cross-sectional area suddenly changes, or where the vertical or rotational
displacement at a point is to be determined.
For example, consider the beam in Fig.1a. Using the same scheme as that for trusses,
four nodes are specified numerically within a circle, and the three elements are identified
numerically within a square. Also, notice that the “near” and “far” ends of each element are
identified by the arrows written alongside each element.

Figure 1:

1.2.Global and Member Coordinates

The global coordinate system will be identified using x , y , z axes that generally have
their origin at a node and are positioned so that the nodes at other points on the beam all have
positive coordinates, Fig.1a. The local or member x, y, z coordinates have their origin at the
“near” end of each element, and the positive axis is directed towards the “far” end. Figure 1b
shows these coordinates for element 2. Notice that for each beam element the x and x axes
will be collinear and the global and member coordinates will all be parallel. Therefore, unlike
the case for trusses, here we will not need to develop transformation matrices between these
coordinate systems.

1.3.Kinematic Indeterminacy

34
 Structural Mechanics For Civil Engineers By Dr. Djomo Mbong

Once the elements and nodes have been

identified, and the global coordinate system has been
established, the degrees of freedom for the beam and
its kinematic determinacy can be determined.
If we consider the effects of both bending and
shear, then each node on a beam can have two
degrees of freedom, namely, a vertical displacement
and a rotation. As in the case of trusses, these linear
and rotational displacements will be identified by
code numbers. The lowest code numbers will be used
to identify the unknown displacements
(unconstrained degrees of freedom), and the highest
numbers are used to identify the known Figure 2:
displacements (constrained degrees of freedom).
To show an example of code-number labeling,
consider again the continuous beam in Fig.1a. Here
the beam is kinematically indeterminate to the fourth
degree. There are eight degrees of freedom, for which
code numbers 1 through 4 represent the unknown
displacements, and numbers 5 through 8 represent the
known displacements, which in this case are all zero.
As another example, the beam in Fig.2a can
be subdivided into three elements and four nodes. In
particular, notice that the internal hinge at node 3
deflects the same for both elements 2 and 3; however, Figure 3:
the rotation at the end of each element is different. For
this reason three code numbers are used to show these deflections. Here there are nine degrees
of freedom, five of which are unknown, as shown in Fig.2b, and four known; again they are all
zero.
Finally, consider the slider mechanism used on the beam in Fig.3a. Here the deflection
of the beam is shown in Fig.3b, and so there are five unknown deflection components labeled
with the lowest code numbers. The beam is kinematically indeterminate to the fifth degree.

Development of the stiffness method for beams follows a similar procedure as that used
for trusses. First we must establish the stiffness matrix for each element, and then these matrices
are combined to form the beam or structure stiffness matrix. Using the structure matrix
equation, we can then proceed to determine the unknown displacements at the nodes and from
this determine the reactions on the beam and the internal shear and moment at the nodes.

2. Beam-Member Stiffness Matrix

In this section we will develop the stiffness matrix for a beam element or member having
a constant cross-sectional area and referenced from the local x, y, z coordinate system, Fig.4.
The origin of the coordinates is placed at the “near” end N , and the positive x axis extends

35
 Chapter Three: Matrix Methods for Beam Analysis

Figure 4:
toward the “far” end F . There are two reactions at each end of the element, consisting of shear
forces q Ny and q Fy and bending moments qNz and qFz . These loadings all act in the positive
coordinate directions. In particular, the moments qNz and qFz are positive counterclockwise.
Linear and angular displacements associated with these loadings also follow this same
positive sign convention. We will now impose each of these displacements separately and then
determine the loadings acting on the member caused by each displacement.

2.1. y  Displacements

When a positive displacement d Ny  is imposed while other possible displacements are

prevented, the resulting shear forces and bending moments that are created are shown in Fig.5a.
The expressions of these forces and bending moments have been developed in Eq.(17) of
chapter 4. Likewise, when d Fy is imposed, the required shear forces and bending moments
are given in Fig.5b.

Figure 5:

2.2. z Rotations

If a positive rotation d Nz is imposed while all other possible displacements are
prevented, the required shear forces and moments necessary for the deformation are shown in

Figure 6:

36
 Structural Mechanics For Civil Engineers By Dr. Djomo Mbong

Fig.6a (See also Eq.(17) of chapter 4). Likewise, when d Fz is imposed, the resultant loadings
are shown in Fig.6b.
By superposition, if the above results in Figs.5 and 6 are added, the resulting four load-
displacement relations for the member can be expressed in matrix form as
Ny Nz  Fy Fz 
 12 EI 6 EI 12 EI 6 EI 
 L3 
L2 L3 L2 
 q Ny    d 
   6 EI 4 EI 6 EI 2 EI   Ny 
 Nz   L2
q L
 2
L L   d Nz 
1
 q    12 EI 6 EI 12 EI 6 EI   d 
 Fy    3  2  2   Fy 
 qFz   L L L3 L  d  
 Fz 
 6 EI 2 EI 6 EI 4 EI 
  2 
 L2 L L L 
These equations can also be written in abbreviated form as
 q   k  d   2
The symmetric matrix  k  in Eq.(1) is referred to as the member stiffness matrix. The 16
influence coefficients kij that comprise it account for the shear-force and bending-moment
displacements of the member. Physically these coefficients represent the load on the member
when the member undergoes a specified unit displacement. For example, if d Ny  1 in Fig.5a,
while all other displacements are zero, the member will be subjected only to the four loadings
indicated in the first column of the  k  matrix. In a similar manner, the other columns of the
 k  matrix are the member loadings for unit displacements identified by the degree-of-freedom
code numbers listed above the columns. From the development, both equilibrium and
compatibility of displacements have been satisfied. Also, it should be noted that this matrix is
the same in both the local and global coordinates since the axes are parallel x , y , z and,
therefore, transformation matrices are not needed between the coordinates.

3. Beam-Structure Stiffness Matrix

Once all the member stiffness matrices have been found, we must assemble them into
the structure stiffness matrix  K  . This process depends on first knowing the location of each
element in the member stiffness matrix. Here the rows and columns of each  k  matrix (Eq.(1))

 
are identified by the two code numbers at the near end of the member N y , N z followed by

 
those at the far end F y , Fz . Therefore, when assembling the matrices, each element must

be placed in the same location of the  K  matrix. In this way,  K  will have an order that will
be equal to the highest code number assigned to the beam, since this represents the total number
of degrees of freedom. Also, where several members are connected to a node, their member

37
 Chapter Three: Matrix Methods for Beam Analysis

stiffness influence coefficients will have the same position in the  K  matrix and therefore
must be algebraically added together to determine the nodal stiffness influence coefficient for
the structure. This is necessary since each coefficient represents the nodal resistance of the
structure in a particular direction ( y or z  ) when a unit displacement ( y or z  ) occurs either
at the same or at another node. For example, K 23 represents the load in the direction and at the
location of code number “2” when a unit displacement occurs in the direction and at the location
of code number “3.”

4. Application of the Stiffness Method for Beam Analysis

After the structure stiffness matrix is determined, the loads at the nodes of the beam can
be related to the displacements using the structure stiffness equation
Q   K  D
Here  Q  and  D  are column matrices that represent both the known and unknown loads and
displacements. Partitioning the stiffness matrix into the known and unknown elements of load
and displacement, we have
 Qk   K 11 K 12   Du 
   
 Qu   K 21 K 22   Dk 
which when expanded yields the two equations
Qk    K 11   Du    K 12   Dk   3
Qu    K 21   Du    K 22   Dk   4
The unknown displacements  Du  are determined from the first of these equations. Using these

values, the support reactions Qu  are computed for the second equation.

4.1. Intermediate Loadings

For application, it is important that the elements of the beam be free of loading along its
length. This is necessary since the stiffness matrix for each element was developed for loadings
applied only at its ends. (See Fig.4.) Oftentimes, however, beams will support a distributed
loading, and this condition will require modification in order to perform the matrix analysis.
To handle this case, we will use the principle of superposition. To show its application,
consider the beam element of length L in Fig.7a, which is subjected to the uniform distributed
load w . First we will apply fixed-end moments and reactions to the element, which will be
used in the stiffness method, Fig.7b. We will refer to these loadings as a column matrix
  q0  . Then the distributed loading and its reactions q0  are applied, Fig.7c. The actual
loading within the beam is determined by adding these two results.

4.2. Member Forces

38
 Structural Mechanics For Civil Engineers By Dr. Djomo Mbong

Figure 7:
The shear and moment at the ends of each beam element can be determined using Eq.(2)
and adding on any fixed-end reactions q0  if the element is subjected to an intermediate
loading. We have:
 q    k  d   q0   5
If the results are negative, it indicates that the loading acts in the opposite direction to that
shown in Fig.4.

Example 1: Determination of the reactions at the supports of the beam shown in Fig.8a.
EI is constant.

Figure 8:

Notation: The beam has two elements and three nodes, which are identified in Fig.8b. The code
numbers 1 through 6 are indicated such that the lowest numbers 1–4 identify the
unconstrained degrees of freedom. The known load and displacement matrices are
 0 1
 
5 2 0 5
Qk     ;  Dk    
 0 3 0 6
 
 0 4

39
 Chapter Three: Matrix Methods for Beam Analysis

Member Stiffness Matrices: Each of the two member stiffness matrices is determined from
Eq.(1). Note carefully how the code numbers for each column and row are established.
6 4 5 3 5 3 2 1
 1.5 1.5 1.5 1.5  6  1.5 1.5 1.5 1.5  5
 1.5 2 1.5 1  4  1.5 2 1.5 1  3
k1   EI  ; k2   EI 
 1.5 1.5 1.5 1.5  5  1.5 1.5 1.5 1.5  2
   
 1.5 1 1.5 2  3  1.5 1 1.5 2  1
Structure Stiffness Matrix : We can now assemble these elements into the structure stiffness
matrix.
1 2 3 4 5 6
 2 1.5 1 0 1.5 0  1
 
 1.5 1.5 1.5 0 1.5 0  2
 1 1.5 4 1 0 1.5  3
 K   k1   k2   EI  0 2 1.5

 0 1 1.5  4
 1.5 1.5 0 1.5 3 1.5  5
 
 0 0 1.5 1.5 1.5 1.5  6
Displacements and Loads. We use Eqs.(3)-(4):

 0  2 1.5 1 0 1.5 0   D1 
     
 5   1.5 1.5 1.5 0 1.5 0   D2 
 0  1 1.5 4 1 0 1.5   D3 
   EI    
 0  0 0 1 2 1.5 1.5   D4 
Q   1.5 1.5 0 1.5 3 1.5   0
 5    
 Q6   0 0 1.5 1.5 1.5 1.5   0 
The matrices are partitioned. Carrying out the multiplication for the first four rows, we have:

 2 D1  1.5 D2  D3  0
  16.67 6.67
 1.5 D1  1.5 D2  1.5 D3   5 D
 1   ; D3  
EI EI
 EI  
 D1  1.5 D2  4 D3  D4  0  D   26.67 ; D4 
3.33
  2 EI EI
 D3  2 D4  0

Using these results, and multiplying the last two rows, gives

1.5 EI  D1  D2  D4   Q5
  Q5  10kN
  
1.5 EI  D3  D4   Q6
  Q6  5kN

40
 Structural Mechanics For Civil Engineers By Dr. Djomo Mbong

Example 2: The beam in Fig.9a is subjected to the two couple moments. If the center support
② settles 1.5 mm, determine the reactions at the supports. Assume the roller supports at ➀
6
and ③ can pull down or push up on the beam. Take E  200GPa and I  22  10 m .
4

Figure 9:
Notation. The beam has two elements and three unknown degrees of freedom. These are labeled
with the lowest code numbers, Fig.9b. Here the known load and displacement matrices are:
 4 1  0  4
   
Qk    0  2;  Dk    0.0015  5
 4  3  0  6
Member Stiffness Matrices: Each of the two member stiffness matrices is determined from
Eq.(1).
6 3 5 2 5 2 4 1
 1.5 1.5 1.5 1.5  6  1.5 1.5 1.5 1.5  5
 1.5 2 1.5 1  3  1.5 2 1.5 1  2
k1   EI  ; k2   EI 
 1.5 1.5 1.5 1.5  5  1.5 1.5 1.5 1.5  4
   
 1.5 1 1.5 2  2  1.5 1 1.5 2  1
Displacements and Loads. We use Eqs.(3)-(4):
1 2 3 4 5 6
 4  2 1 0 1.5 1.5 0  D1 
     
 0  1 4 1 1.5 0 1.5   D2 
 4   0 1 2 0 1.5 1.5   D3 
   EI    
 Q4   1.5 1.5 0 1.5 1.5 0   0 
Q   1.5 0 1.5 1.5 3 1.5   0.0015 
 5    
 Q6   0 1.5 1.5 0 1.5 1.5   0 

The matrices are partitioned. Carrying out the multiplication for the first three rows, we have:
 EI  2 D1  D2  1.5  0.0015    4  D1  0.00158rad
 
 D1  4 D2  D3  0   D2  0
 
 EI  D2  2 D3  1.5  0.0015    4  D3  0.00158rad

41
 Chapter Three: Matrix Methods for Beam Analysis

Using these results, and multiplying the last three rows, gives
Q4  0.525kN ; Q5  1.05kN ; Q6  0.525kN

Example 3: Determination of the moment developed at support A of the beam shown in

Fig.10a. Assume the roller supports can pull down or push up on the beam. Take
E  29  103 ksi and I  510in4 .

note:1152kin12kft=1008kin.

Figure 10:
Notation. Here the beam has two unconstrained degrees of freedom, identified by the code
numbers 1 and 2.
The matrix analysis requires that the external loading be applied at the nodes, and therefore
the distributed and concentrated loads are replaced by their equivalent fixed-end moments,
which are determined from the table on chapter 4. Note that no external loads are placed at ①
and no external vertical forces are placed at ② since the reactions at code numbers 3, 4 and
5 are to be unknowns in the load matrix. Using superposition, the results of the matrix analysis
for the loading in Fig.10b will later be modified by the loads shown in Fig.10c. From Fig.10b,
the known displacement and load matrices are:
0 4
 144  1
Qk     ;  Dk    0  5
 1008 2  0 6
Member Stiffness Matrices. Each of the two member stiffness matrices is determined from
Eq.(1).
4 3 5 2
 7.430 1069.9 7.430 1069.9  4
 1069.9 205417 1069.9 102708  3
k1     ;
 7.430 1069.9 7.430 1069.9  5
 
 1069.9 102708 1069.9 205417  2

42
 Structural Mechanics For Civil Engineers By Dr. Djomo Mbong

5 2 6 1
 200.602 9628.91 200.602 9628.91  5
 9628.91 616250 9628.91 308125  2
k2   
 200.602 9628.91 200.602 9628.91 6
 
 9628.91 308125 9628.91 616250  1
Displacements and Loads. We require
Q   K  D
 144   616250 308125 0 0 9628.91 9628.91   D1 
     
 1008   308125 821667 102708 1069.9 8559.01 9628.91   D2 
 Q3   0 102708 205417 1069.9 1069.9 0   0
   EI    
 Q4   0 1069.9 1069.9 7.430 7.430 0   0
 Q   9628.91 8559.01 1069.9 7.430 208.03 200.602   0
 5
    
 Q6   9628.91 9628.91 0 0 200.602 200.602   0 
Solving in the usual manner,
 616250 D1  308125 D2  144  3
 D1  0.4673  10 in
  
 308125 D1  821667 D2  1008
3
 D2  1.40203  10 in

Thus,
Q3  0  102708  1.40203  103  144k  in  12k  ft
The actual moment at A must include the fixed-supported reaction of 96k  ft shown in Fig.10c,
along with the calculated result for Q 3 . Thus,
M AB  12k  ft  96k  ft  108 k  ft

43