Multi-Degree of Freedom Systems

Vibration

Delbante Alebachew
 Chapter 4 Multiple Degree of Freedom Systems
▪ If the real dynamic behavior of systems is needed to be
studied with reasonable accuracy, a model with finite
number of DoF has to be used.
• This is true since all structural systems are continuous systems
with an infinite number of DoF.
The first step in analyzing multiple degrees of
freedom (DOF) is to look at 2 DOF
• DOF: Minimum number of coordinates to specify the position
and orientation of a system
• Many systems have more than 1 DOF
• Examples of 2 DOF systems
– car with spring (heave)
– elastic pendulum (radial and angular)
– motions of a ship (roll and pitch)

Fig 4.1
4.1 Two-Degree-of-Freedom Model (Undamped)

A two degree of freedom system used to base much

of the analysis and conceptual development of
MDOF systems on.
 Free-Body Diagram of each mass

k2(x2 -x1)

k 1 x1 m1 m2

x1 x2

Figure 4.2
Summing forces yields the equations of motion:
m1 x1 (t ) = −k1 x1 (t ) + k2 ( x2 (t ) − x1 (t ) )
(4.1)
m2 x2 (t ) = −k2 ( x2 (t ) − x1 (t ) )

Rearranging terms:
m1 x1 (t ) + (k1 + k2 ) x1 (t ) − k2 x2 (t ) = 0
(4.2)
m2 x2 (t ) − k2 x1 (t ) + k2 x2 (t ) = 0

This is the matrix/vector representation of Eqn. (4.1). The mass and

stiffness matrices, M and K, described above have the special
property of being symmetric.
Symmetric matrix is a matrix that is equal to its transpose.
 Note
• A two Degree-of-Freedom system has
– Two equations of motion! (homogeneous & coupled eqs.)
• If free vibrations, so homogeneous eqs.
• Equations are coupled:
– Both have x1 and x2.
– If only one mass moves, the other follows
– Example: pitch and heave of a car model
• In this case the coupling is due to k2.
– Mathematically and Physically
– If k2 = 0, no coupling occurs and can be solved as two
independent SDOF systems
– Two natural frequencies!
 Initial Conditions

• Two coupled, second-order, ordinary differential

equations with constant coefficients
• Needs 4 constants of integration to solve
• Thus 4 initial conditions on positions and velocities

x1 (0) = x10 , x1 (0) = x10 , x2 (0) = x20 , x2 (0) = x20

• The initial conditions can also be written in terms of
vectors as:

 x1 (0)   x1 (0) 
xo =   and xo =   .........................
 x2 (0)   x2 (0)
Solution by Matrix Methods

(4.4)

 x1 (t )   x1 (t )   x1 (t ) 
x(t ) =   , x(t ) =   , x(t ) =   (4.5)
 2 
x (t )  2 
x (t )  2 
x (t )

m 0  k + k −k2 
M = 1  K = 1 2
k2 
, (4.6), (4.7)
 0 m2   −k2

Mx + Kx = 0 m1x1 (t ) + (k1 + k2 ) x1 (t ) − k2 x2 (t ) = 0
m2 x2 (t ) − k2 x1 (t ) + k2 x2 (t ) = 0
 The approach to a Solution:
For 1DOF (Ch1 & Ch2) we assumed the scalar solution Aet
Similarly, now we assume the vector form:
jt
Let x(t ) = ue (4.8)

j = −1, u = 0, , u unknown
❑
 -(
Substitution of this
M + K ue jt
=)
2 assumed solution
0 into the vector
equation of motion given by M x + Kx = 0 ......................................
yields
 ( - M + K ) u = 0
2
(4.15)
This changes the differential
equation of motion into algebraic
vector equation:

( - M + K ) u = 0
2
(4.17)
This is two algebraic equation in 3 uknowns
( 1 vector of two elements and 1 scalar):
 u1 
u=   , and 
u2 
The condition for solution of this matrix
equation requires that the matrix inverse
does not exist:
✓ If. the inv ( - M + K ) exists  u = 0: which is the
2

static equilibrium position. Means

For motion to occur
no motion

 ( -toMoccur
+ K ) does not exist
−1
= 0motion
✓ uFor
2

or det ( - 2 M + K ) = 0 (4.19)

The determinant results one equation (called the

characteristic equation) with one unknown ()
Back to our specific system: the
characteristic equation is defined as
det(- M + K )= 0 
2

− 2 m1 + k1 + k2 −k2 
det =0 
 −k2 − m2 + k2 
2

m1m2 − (m1k2 + m2 k1 + m2 k2 ) + k1 k2 = 0
4 2

 and   1 and  2

2
1
2
2
Once  is known, use equation again
to calculate the corresponding vectors u1 and u2
This yields vector equation for each squared frequency:
(−12 M + K )u1 = 0 (4.22)
and
(−22 M + K )u2 = 0 (4.23)

❑ Each of these matrix equations represents 2

equations in the 2 unknowns components of the
vector, but
❑ The coefficient matrix is singular so each matrix
equation results in only 1 independent equation.
Examples 1: Calculating u and 

m1= 9 kg, m2=1kg, k1=24 N/m and k2=3 N/m

The characteristic equation becomes
4 - 62 + 8 = (2 - 2) (2 - 4) = 0
2 = 2 and 2 = 4 or
1,3 =  2 rad/s, 2,4 =  2 rad/s
Each value of 2 yields an expression or u:
 Computing the vectors u
u11 
For  =2, denote u1 =   then we have
2
1
u12 
(-12 M + K )u1 = 0 
 27 − 9(2) −3  u11  0
 −3   = 
 3 − (2)  u12  0
9u11 − 3u12 = 0 and − 3u11 + u12 = 0
2 equations, 2 unknowns but DEPENDENT!
(the 2nd equation is -3 times the first)
Only the direction of vectors u can be
determined, not the magnitude as it
remains arbitrary
u11 1 1
=  u11 = u12 results from both equations:
u12 3 3
only the direction, not the magnitude can be determined!
This is because: det(−12 M + K ) = 0.
The magnitude of the vector is arbitrary. To see this suppose
that u1 satisfies
(−12 M + K )u1 = 0, so does au1 , a arbitrary. So
(−12 M + K )au1 = 0  (−12 M + K )u1 = 0
 u21 
For  = 4, let u2 =   then we have
2
2
u22 
(-1 M + K )u = 0 
2

 27 − 9(4) −3  u21  0

 −3   = 
 3 − (4)  u22  0
1
−9u21 − 3u22 = 0 or u21 = − u22
3
Note that the other equation is the same
What to do about the magnitude!
Several possibilities, here we just fix one element:

Choose:
 3 1
u12 = 1  u1 =  
1
Choose:
 3
−1
u22 = 1  u2 =  
1
Thus the solution to the algebraic matrix equation is:

 13
1,3 =  2, has mode shape u1 =  
1
 −1 3 
 2,4 =  2, has mode shape u2 =  
1

Here we have introduce the name

mode shape to describe the vectors u1 and u2.
The origin of this name comes later
Return now to the time response:
We have computed four solutions:
x(t ) = u1e− j1t , u1e j1t , u2e− j2t , u2e j2t 
Since linear, we can combine as:
− j1t j1t − j2t j2t
x(t ) = au1e + bu1e + cu2e + du2e
 x(t ) = ( ae− j1t + be j1t ) u1 + ( ce− j2t + de j2t ) u2
= A1 sin(1t + 1 )u1 + A2 sin(2t + 2 )u2

where A1 , A2 , 1 , and 2 are constants of integration

Note that to go from the exponential
determined by initial conditions. form to to sine requires Euler’s
formula for trig functions and uses up
the +/- sign on omega
Physical interpretation of all that math!

x(t)
• Each of the TWO masses is oscillating at TWO
natural frequencies 1 and 2

• The relative magnitude of each sine term, and

hence of the magnitude of oscillation of m1 and m2
is determined by the value of A1u1 and A2u2

• The vectors u1 and u2 are called mode shapes

because the describe the relative magnitude of
oscillation between the two masses
What is a mode shape?
• First note that A1, A2, 1 and 2 are determined by
the initial conditions
• Choose them so that A2 = 1 = 2 =0
• Then:
 x1 (t)  u11 
x(t) =   = A1   sin 1t = A1u1 sin 1t
 x2 (t) u12 
• Thus each mass oscillates at (one) frequency 1
with magnitudes proportional to u1 the 1st mode
shape
A graphic look at mode shapes:
If initial condition’s correspond to mode 1 or 2, then the response is
purely in mode 1 or mode 2.

x1 x2
k1 k2
 13
Mode 1: m1 m2 u1 =  
1
x1=A/3 x2=A

x1 x2
k1
 −1 3 
k2
Mode 2: m1 m2 u2 =  
1
x1=-A/3 x2=A
Example 4.1 given the initial conditions
compute the time response

1 0
consider x(0) =   mm, x (0) =  
0 0
 x1 (t )   sin ( 2t + 1 ) − 2 sin (2t + 2 )
 A1 A
 x (t ) =  3 3
 2   A1 sin ( 2t + 1 ) + A2 sin (2t + 2 ) 
 2 cos( 2t + 1 ) − 2 cos(2t + 2 )

A A
 x1 (t )   1 2

 x (t ) =  3 3
 2   A1 2 cos( 2t + 1 ) + A2 2 cos(2t + 2 ) 
At t = 0 we have

 
1 mm 
= 
A1
sin ( ) −
A2
sin ( ) 
 0 
1 2
3 3
   A1 sin (1 ) + A2 sin (2 ) 
 
2 cos(1 ) − 2 cos(2 )
A1 A2
0 
0 =  3 3

 A1 2 cos(1 ) + 2 A2 cos(2 ) 
 
4 equations in 4 unknowns:

3 = A1 sin(1 ) − A2 sin (2 )

0 = A1 sin (1 ) + A2 sin(2 )
0 = A1 2 cos(1 ) − A2 2 cos(2 )
0 = A1 2 cos(1 ) + A2 2 cos(2 )
Yields:

A1 = 1.5 mm, A2 = −1.5 mm, 1 = 2 = rad
2
The final solution is:
x1 (t) = 0.5cos 2t + 0.5cos2t
x2 (t) = 1.5cos 2t − 1.5cos2t
These initial conditions gives a response that is a combination
of modes. Both harmonic, but their summation is not.

Figure 4.3
Solution as a sum of modes

x(t) = a1u1 cos1t + a2u2 cos2t

Determines how the second

frequency contributes to the
Determines how the first response
frequency contributes to the
response
Things to note
• Two degrees of freedom implies two natural
frequencies
• Each mass oscillates at with these two frequencies
present in the response and beats could result
• Frequencies are not those of two component
systems
k1 k2
1 = 2  = 1.63, 2 = 2  = 1.732
m1 m2

• The above is not the most efficient way to

calculate frequencies as the following describes
Some matrix and vector reminders
a b −1 1  d −b 
A=  A =  −c a 
 d
c c ad − cb  
xT x = x12 + x22
 m1 0 
M =   xT
Mx = m x
1 1
2
+ m x
2 2
2

 0 m2 
M  0  x Mx  0 for every value of x except 0
T

Then M is said to be positive definite

4.2 Eigenvalues and Natural Frequencies
▪ The algebraic eigenvalue problem allows the power of
mathematics to be used in solving vibration problems.
▪ This will give us some powerful computational skills
and some powerful theory
▪ All the codes have eigen-solvers so these painful
calculations can be automated
▪ Sets the background needed for analyzing systems with
arbitrary number of degree of freedom.
▪ In addition, the important concepts of mode shapes and
natural frequencies can be generalized
Some matrix results to help us use available
computational tools:

▪ A matrix M is defined to be symmetric if

M=M T

▪ A symmetric matrix M is positive definite if

x Mx  0 for all nonzero vectors x
T

▪ A symmetric positive definite matrix M can be factored

M = LL T

Here L is upper triangular, called a Cholesky matrix

Some matrix results to help us use available
computational tools:

▪ If the matrix L is diagonal, the matrix square root is root

of the diagonal element

The matrix square root is the matrix M 1/2 such that

M M =M
1/2 1/2

If M is diagonal, then the matrix square root is just the root

of the diagonal elements:
 m1 0 
L=M 1/2
=  (4.35)
 0 m2 
A change of coordinates is introduced to capitalize
on existing mathematics

For a symmetric, positive definite matrix M:

m1 0  −1  1 m1 0 −1/ 2
 1
m1
0 
M =  , M = 1 
, M = 
 0 m2  0 m2   0 1
m2 

Let x(t ) = M −1/ 2q(t ) and multiply by M −1/ 2 :
M −1/ 2 MM −1/ 2 q(t ) + M −1/ 2 KM −1/ 2 q(t ) = 0 (4.38)
I identity K symmetric

or q(t ) + Kq(t ) = 0 where K = M −1/ 2 KM −1/ 2

k
K is called the mass normalized stiffness and is similar to the scalar
m
used extensively in single degree of freedom analysis. The key here is that
K is a SYMMETRIC matrix allowing the use of many nice properties and
computational tools
Solution of vibration problem relates to the real
symmetric eigenvalue problem
jt
Assume q(t ) = ve in q(t ) + Kq(t ) = 0
− ve2 jt
+ Kve jt
= 0, v  0 or
Kv =  v 2
 Kv =  v v0
vibration problem real symmetric
eigenvalue problem
(4.40) (4.41)
Important Properties of the n x n Real
Symmetric Eigenvalue Problem
Square Matrix Review
Normal and orthogonal vectors

 x1   y1  n
   
x =  M , y =  M , inner product is x y =  xi yi
T

i =1
 xn   yn 
x orthogonal to y if xT y = 0 

x is normal if x x = 1
T

if a the set of vectores is is both orthogonal and normal it
is called an orthonormal set

The norm of x is x = xT x =1

V (eigenvectors are both orthogonal and normal)

 Normalizing any vector can be done by dividing
it by its norm:

x
T
has norm of 1
x x

To see this compute

x xT x xT x
= = T =1
T
x x T T
x x x x x x
Examples

−1/ 2 −1/ 2  13 0 27 −3  13 0

K = M KM =   −3 3   0 1
 0 1   
 3 −1
so K =   which is symmetric.
−1 3 
3- -1 
det(K −  I ) = det   =  2
− 6 + 8 = 0
 -1 3- 
which has roots: 1 = 2 = 12 and 2 = 4 = 22
( K − 1I ) v1 = 0 
3 − 2 −1   v11  0
 −1 3 − 2 v  = 0 
   12   
1
v11 − v12 = 0  v1 =   
1
v1 =  (1 + 1) = 1   =
2 1
2

1 1 The first normalized eigenvector

v1 = 1
2 
Likewise the second normalized eigenvector is
computed and shown to be orthogonal to the first,
so that the set is orthonormal
1 1 1
v2 =  −1 , v1 v 2 = 2 (1 − 1) = 0
T

2 
1
v1 v1 = (1 + 1) = 1
T

2
1
v 2 v 2 = (1 + (−1)(−1)) = 1
T

2
 v i are orthonormal
Modes u and Eigenvectors v are different but
related:

u1  v1 and u2  v 2
−1/2 −1/2
x=M qu= M v
Note
 3 0   1  1
M u1 = 
1/2
 3 = = v1
  
 0 1   1  1
This orthonormal set of vectors is used to form
an Orthogonal Matrix

P =  v1 v 2  called a matrix of eigenvectors (normalized)

 vT
v v T
1 v2
 1 0 
P P= T
T 1 1
T =  =I
 v 2 v1 v 2 v 2  0 1
P is called an orthogonal matrix
PT KP = PT  Kv1 Kv 2  = PT 1v1 2 v 2 
1v1T v1 2 v1T v 2  1 0 
= T =
   = diag( 2
, 2)=
2

1v 2 v1 2 v 2 v 2   0 2 
T 1

~
 = diag (i ) = P KP is also called a modal matrix
T
Modal Analysis
▪ Consider the matrix form of equation of vibration
Mx + Kx = 0 (8.31)
▪ Subject to the initial conditions of
x(0) = x0 ,  (0) = x
x  0 , where
x0 = x1 (0) x2 (0)
T

 0 = x1 (0)
x x2 (0)
T

− 12
▪ Substitution of x = M q into equation and multiplying
− 12
from the left by Μ yields:
~
(t ) + Kq(t ) = 0
Iq (8.32)

− 12 ~ − 12 − 12
x = M  and K = Μ KΜ
q
46
 Modal Analysis
• The transformation Μ− 12 simply transforms the problem from
the coordinate system defined by
x = x1 (t ) x2 (t )
T

to a new coordinate system defined by

q = q1 (t ) q2 (t )
T

• The initial conditions in the new coordinate system will be:

1 1
q(0) = Μ x0 and q (0) = Μ 2 x 0
2
(8.33)
Next define a second coordinate system r = r1(t) r2 (t)T
by: q(t ) = Pr(t )

~
(t ) + Kq(t ) = 0
• Substitution of the vector q(t ) = Pr(t ) into Iq
T
and multiplying from the left by matrix P yields:
~
P Pr(t ) + P KPr (t ) = 0
T T

~
• Using the result PT P = I and  = PT KP = diag (i ), will be
reduced to:

Ir(t ) + r(t ) = 0
• Equation Ir(t ) + r(t ) = 0 can be written out by performing
the indicated matrix calculations as:

1 0  r1 (t )  12 0   r1 (t )  0

0 1 r (t ) +  2   = 
   2   0 2  r2 (t ) 0
 r1 (t ) + 12r1 (t )  0
 = 
r2 (t ) + 2 r2 (t ) 0
2

• The equality of the two vectors in this last expression implies

the two decoupled equations

r1 (t ) + 12r1 (t ) = 0
and
r2 (t ) +  r (t ) = 0
2
2 2
• The coordinate system r = r1 (t ) r2 (t ) is called the modal
T

coordinate system.
• Hence each equation can be solved independently by using
the same method as in the previous chapters.

12r102 + r102  −1 1r10 

r1 (t ) = sin  1t + tan 
1  r10 
22r202 + r202  −1 2 r20 
r2 (t ) = sin  2t + tan 
2  r20 
• Denoting the initial conditions individually by r10 , r10 , r20 , & r20 ,
the solution of each of the modal equations is simply
The transformed initial conditions become

q = Pr ,  r = PT q

1 1
q(0) = Μ x0 and q (0) = Μ 2 x 0
2
(8.33)

 r10  1 
r(0) =   = P q(0) = P M x0  Initial displacement
T T 2

r20  

 r10 
r (0) =   = P q (0) = P M 2 x 0 
1
T T
Initial velocity
 
 
r 20 
• Then to obtain the vector x from vector r,
−1 −1
x = M q = M 2 Pr(t)
2
−1
• The matrix product M 2 P is again a matrix, which is
denoted by
−1
S =M 2
P
• It is the same size as M and P and used to reduce the number
of steps required to perform the procedure indicated above.

52
• In order to obtain the solution in the original physical
coordinate system, x(t), the transformation S is again
used. Multiplying r(t) = S-1x(t) by S to get

−1
x(t ) = Sr(t ) = M 2
P r(t) (8.5

12r102 + r102  −1 1r10 

r1 (t ) = sin  1t + tan 
1  r10 
22r202 + r202  −1 2 r20 
r2 (t ) = sin  2t + tan 
2  r20 
 The following summarizes the procedure of
modal analysis using the matrix transformation S.
−1
• Taking the transpose of eqn S = M 2
P and knowing
that (AB)T = BTAT, we will get:
−1 −1
S = (M
T 2
P) = P M
T T 2
(8.46)
• In addition, the inverse of a matrix product is given by
(AB)-1 = B-1A-1, so that
−1 1
−1 −1 −1
S = (M 2
P) = P M 2
(8.47)
• However, the matrix P has as its inverse PT since PTP = I,
thus 1
S −1 = PT M 2
(8.48)

54
 Mx + Kx = 0

ST MSr(t ) + ST KSr(t ) = 0 (8.49)

−1
S =M 2
P
−1 −1 −1 −1
T
P M 2
MM 2
Pr(t ) + P M
T 2
KM 2
Pr(t ) = 0 (8.50)
or
~
PT Pr(t ) + PT KPr(t ) = 0 (8.51)

r(t ) + r(t ) = 0 (8.52)

55
• The initial conditions on r(t) are calculated by solving for r(t)
in −1 −1
x=M 2
q=M 2
Pr(t) (8.44)
• Multiplying by S-1 yields r(t) = S-1x(t), which becomes,
1
−1
r(t ) = S x(t ) = P M 2 x(t )
T
(8.53)
• The initial conditions on r(t) are thus
1 1
r(0) = P M x0 and r (0) = P M 2 x 0
T 2 T
(8.54)
• In order to obtain the solution in the original physical
coordinate system, x(t), the transformation S is again used.
Multiplying r(t) = S-1x(t) by S to get
−1
x(t ) = Sr(t ) = M 2
P r(t) (8.55)
Example: compute P and show that it is an
orthogonal matrix

From the previous example:

1 1 1 
P = v1 v1 = 1 −1 
2 
1 1 1 1  1 1 
P P=
T
1 −1 1 −1
2 2  
1 1 + 1 1 − 1 1  2 0 
=   =   =I
2 1 − 1 1 + 1 2  0 2 
Example: Compute the square of the frequencies
by matrix manipulation

1 1 1   3 −1 1 1 1 
P KP =
T
    
2 1 −1 −1 3  2 1 −1
1 1 1   2 4 
= 
2 1 −1  2 −4
1 4 0 2 0  12 0 
=   =  == 2
2 0 8   0 4   0 2 
 1 = 2 rad/s and 2 = 2 rad/s
In general:
 = PT KP = diag ( i ) = diag(i2 ) (4.48)
▪ The next step is to calculate the matrix S and its inverse.

▪ The modal initial conditions are calculated from equations

▪ The solution in the physical coordinate system x(t) is calculated from

Example:

m1 = 1 kg, m2 = 4 kg, k1 = k3 = 10 N/m and k2 =2 N/m

The equations of motion:
m1 x1 + (k1 + k2 ) x1 − k2 x2 = 0
(4.49)
m2 x2 − k2 x1 + (k2 + k3 ) x2 = 0
In matrix form these become:

m1 0  k1 + k2 −k 2 
 0 m  x +  −k 
k2 + k3 
x=0 (4.50)
 2  2

Calculate eigenvector
Next substitute numerical values
and compute P and 
m1 = 1 kg, m2 = 4 kg, k1 = k3 = 10 N/m and k2 =2 N/m
1 0  12 −2
M =  , K =
0 4  −2 12 
−1/ 2 −1/ 2 12 −1
 K = M KM =
 −1 12 
12 −  −1 
 det ( K −  I ) = det   =  2
− 15 + 35 = 0
 −1 12 −  
 1 = 2.8902 and 2 = 12.1098
 1 = 1.7 rad/s and 2 = 12.1098 rad/s
Next compute the eigenvectors
For 1 equation (4.41 ) becomes:
12 - 2.8902 −1   v11 
    =0
 −1 3- 2.8902 v21 
 9.1089v11 = v21
Normalizing v1 yields
1 = v1 = v112 + v21
2
= v112 + (9.1089)2 v112
 v11 = 0.1091, and v21 = 0.9940
 0.1091  −0.9940
v1 =   , likewise v 2 =  
0.9940  0.1091 
Next check the value of P to see
if it behaves as its suppose to:
 0.1091 −0.9940
P =  v1 v2  =  
0.9940 0.1091 

 0.1091 0.9940 12 −1  0.1091 −0.9940  2.8402 0 

PT KP =    =
 
 −0.9940 0.1091  −1 3  0.9940 0.1091   0 12.1098

 0.1091 0.9940  0.1091 −0.9940 1 0

PT P =     = 
 −0.9940 0.1091 0.9940 0.1091  0 1

Yes!
A note on eigenvectors
In the previous section, we could have chosed v 2 to be
 0.9940  -0.9940 
v2 =   instead of v 2 =  
 −0.1091   0.1091 
because one can always multiple an eigenvector by a constant
and if the constant is -1 the result is still a normalized vector.

Does this make any difference?

No! Try it in the previous example
All of the previous examples can and should
be solved by “hand” to learn the methods
However, they can also be solved on
calculators with matrix functions and
with the codes listed in the last section
In fact, for more then two DOF one must
use a code to solve for the natural
frequencies and mode shapes.
Next we examine 3 other formulations for
solving for modal data
Matlab commands
• To compute the inverse of the square matrix
A: inv(A) or use A\eye(n) where n is
the size of the matrix
• [P,D]=eig(A) computes the eigenvalues
and normalized eigenvectors (watch the
order). Stores them in the eigenvector
matrix P and the diagonal matrix D (D=)
More commands
• To compute the matrix square root use
sqrtm(A)
• To compute the Cholesky factor: L= chol(M)
• To compute the norm: norm(x)
• To compute the determinant det(A)
• To enter a matrix:
K=[27 -3;-3 3]; M=[9 0;0 1];
• To multiply: K*inv(chol(M))
An alternate approach to normalizing
mode shapes
From equation ( )
−M 2 + K u = 0, u0

Now scale the mode shapes by computing  such that

1
(i ui ) M (i ui ) = 1  i = T
T

ui ui

wi = i ui is called mass normalized and it satisfies:

−i2 Mwi + Kwi = 0  i2 = wTi Kwi , i = 1, 2
There are 3 approaches to computing
mode shapes and frequencies
−1 −1
(i)  Mu = Ku (ii)  u = M Ku
2 2 −1
(iii)  v = M KM 2 v
2 2

(i) Is the Generalized Symmetric Eigenvalue Problem

easy for hand computations, inefficient for computers

(ii) Is the Asymmetric Eigenvalue Problem

very expensive computationally

(iii) Is the Symmetric Eigenvalue Problem

the cheapest computationally