0% found this document useful (0 votes)
42 views10 pages

7 5 Second Order Systems and Mechanical Applications

Uploaded by

Arheta Aretha
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
0% found this document useful (0 votes)
42 views10 pages

7 5 Second Order Systems and Mechanical Applications

Uploaded by

Arheta Aretha
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/ 10

MATH 2243: Linear Algebra & Differential Equations

Discussion Instructor: Jodin Morey [email protected]


Website: math.umn.edu/~moreyjc

7.5: Second-Order Systems and Mechanical Applications

Masses (n of them) connected to each other and connected to two walls by n 1 springs. Assume no
friction, and that each mass m j reacts to the spring(s) attached to it by the familiar formula
F mjxj kx. So, assuming the mass in question m j is reacting to two springs (k j and k j 1 ), we
have: F m j x j kj xj xj 1 kj 1 xj 1 xj .

Case: n 3
m1x1 k1x1 k2 x2 x1 ,
m2x2 k2 x2 x1 k3 x3 x2 ,
m3x3 k3 x3 x2 k4x3.
Observe that the initial k 1 and the final spring k n 1 only have one mass displacement effecting it (x 1
and x n , respectively).

We can put the displacement x j of each mass m j into a displacement vector: x x1 x2 x3 .


Similarly with the masses, we have a mass matrix:
m1 0 0
M 0 m2 0 .
0 0 m3
For the spring constants, we have this stiffness matrix:
k1 k2 k2 0
K k2 k2 k3 k3 .
0 k3 k3 k4

Using these mathematical objects, we can more elegantly represent the the above system as
Mx Kx. Since M is invertible, we can calculate M 1 and multiply both sides of the equation (on
the left) to simplify our equation further to our familiar x Ax, where A M 1 K.

Solution of Second-Order Homogeneous Systems: x Ax


Consider solutions of the form e rt , which we used for single equations. To solve for a system,
however, we will need to make this into a vector. Multiplying by a generic constant vector v , we have
v e rt . Assuming a solution of this form, and plugging it back into our DEQ, we get:
A v e rt v e rt r v e rt r 2 v e rt .
Dividing by e rt , we get A v r 2 v . But this is the eigenvector/eigenvalue equation where v is an
eigenvector for A, and r 2 is the associated eigenvalue.

Typically, when systems of equations like these model mechanical systems, we have eigenvalues
2
j j of A which are less than or equal to zero (where each j is a circular frequency). This
2 i jt
gives us r j j j i. So, for the eigenpair j , v j of A we have: v j e cos jt i sin jt v j.
And from the real and imaginary parts, we get: x j t a j cos j t b j sin j t v j .
2 2 2 , with
Theorem: If the n n matrix A has n distinct nonpositive eigenvalues 1, 2, , n
eigenvectors v 1 , v 2 , , v n , then a general solution of x Ax is given by
n
x t a cos j t b j sin j t v j , where a j and b j are arbitrary constants. In the case where
j 1 j
2
j 0, the corresponding part x j t of the general solution is aj bjt v j .
n
We wish to convert the solution above to the form x t c cos j t
j 1 j j v j , where j is the
"phase shift" or "phase angle."
So, recall (or learn for the first time) that if we have: A cos t B sin t.
and wish to alter it to be like: C cos t , (where C turns out to be the amplitude of the
vibration)
we let A and B be the legs of a right triangle. Then the hypotenus is: C A2 B2 .

A B
With angle (opposite of B), recall we have: cos C
, sin C
,
tan 1 B if A, B 0 (1st quadrant),
A
where tan 1 BA if A 0 (2nd/3rd quadrant),
2 tan 1 BA if A 0, B 0 (4th quadrant).

Thus we transform into, A cos t B sin t C CA cos t CB sin t


C cos cos t sin sin t .
Recall the Trigonometric Identity: cos x cos y sin y sin x cos x y cos y x .
So we get: C cos t , where C is the amplitude,
is the circular frequency in radsec , and is the phase angle.
2 1 cycles
Period of Motion: T sec. Frequency: v T 2
in sec .
So returning to x t , we have x j t c j cos j cos 5t sin j sin 5t v j c j cos 5t j v j.

Superposition of Wave Frequecies 1 and 2 :

Here is a video showing the kinds of movements involved in this section:


https://www.youtube.com/watch?v cu4TvUwk17g

Forced Oscillations and Resonance:


T
Let Mx Kx F where F F1 t F2 t Fn t are the external forces acting on the masses
m1, m2, , mn .
1 2 F t
n F t F t T
So, x Ax f , where f m1 m2 m2 is the external force vector per unit mass.
Often the external forces are periodic,
and we have f t F 0 cos t, where F 0 is some constant vector.
We obtain resonance when the external (forced) frequency is equal
to one of the system’s internal frequencies 1 , 2 , , n .
Undetermined coefficients suggests a trial solution of:
x trial t c cos t. (why not " b sin t" ??)
We solve for particular solution by plugging in this trial solution,
and determining the coefficients: c c1 c2 cn .

As with a single equation with forced oscillation, we have a periodic and transient solution
x t x tr t x sp t (see section 5.6). Given any damping, the transient solution eventually
disappears leaving only the periodic solution (which is being induced by the external force).

Problem: #7 Suppose a mass-and-spring system have the following stiffness matrix


k1 k2 k2
K
k2 k2 k3
and has the following values for the mass and spring constants...
m1 m2 1; k1 4, k 2 6, k 3 4.
Find the two natural frequencies of the system and describe its two natural modes of oscillation.

Mx Kx or x M 1 Kx.

m1 0 1 0
M
0 m2 0 1

M 1.

1 0 4 6 6 10 6
So, A .
0 1 6 6 4 6 10

10 6 2 2
10 36 20 64 16 4 .
6 10
Eigenvalues 1 4 and 2 16,
T T
with associated eigenvectors v 1 11 and v 2 1 1 .

n 2
Recall: "x t j 1
a j cos jt b j sin jt v j" and "Eigenvalues: i"

1 1
Therefore: x t a 1 cos 2t b 1 sin 2t a 2 cos 4t b 2 sin 4t
1 1

x1 t a 1 cos 2t b 1 sin 2t a 2 cos 4t b 2 sin 4t,


x2 t a 1 cos 2t b 1 sin 2t a 2 cos 4t b 2 sin 4t.

"Describe its two natural modes of oscillation."

The natural frequencies are 1 2 and 2 4. In the natural mode with frequency 2, the two masses
m 1 and m 2 move in the same direction with equal amplitudes of oscillation. At frequencies 4, they
move in opposite directions with equal amplitudes.

Problem: #10 The mass-and-spring system of the problem #7 (above) is set in motion from rest [
x1 0 x2 0 0 ], at its equilibrium position [ x 1 0 x2 0 0 ], with external forces
F1 t 30 cos t and F 2 t 60 cos t acting on the masses m 1 and m 2 , respectively. Find the resulting
motion of the system and describe it as a superposition of oscillations.
10 6
Recall: x Ax f , m1 1, m 2 1, and A .
6 10

Observe that f M 1F F 30 cos t, 60 cos t (since M I.

So, forming the nonhomogeneous DEQ x Ax f , we have:


x1 10x 1 6x 2 30 cos t,
x2 6x 1 10x 2 60 cos t

Recall complementary solution from prob. 7:


x c,1 t a 1 cos 2t a 2 sin 2t b 1 cos 4t b 2 sin 4t
x c,2 t a 1 cos 2t a 2 sin 2t b 1 cos 4t b 2 sin 4t

Recall from the review that the "trial solution is: x trial t c cos t, " where we can label the
components c : d1 d2 .

Taking derivatives of the of the trial solution x 1 d 1 cos t, x 2 d 2 cos t in order to substitute into
the system :
x1 d 1 sin t, x2 d 2 sin t, x1 d 1 cos t, x2 d 2 cos t.
d 1 cos t 10 d 1 cos t 6 d 2 cos t 30 cos t,
d 2 cos t 6 d 1 cos t 10 d 2 cos t 60 cos t.

Dividing by cos t :
d1 10d 1 6d 2 30 ,
d2 6d 1 10d 2 60 . (two equations in two unknowns)

2 10 2 2 10 20
9d 1 6d 2 30 , 9d 2 6d 1 60 ; d1 3
d2 3
, d2 3 3
d2 3 3
5 80 2 10
9
d2 9
, d2 16, d1 3
16 3
14.

So a general solution is given by:


x1 t a 1 cos 2t a 2 sin 2t b 1 cos 4t b 2 sin 4t 14 cos t,
x2 t a 1 cos 2t a 2 sin 2t b 1 cos 4t b 2 sin 4t 16 cos t.

Initial conditions: x 1 0 x2 0 0
0 a1 b1 14, 0 a1 b1 16;

So: a1 b1 14 , 0 b1 14 b1 16,
2b 1 2, b 1 1; a1 1 14 15.

Now taking the derivative for the initial condition: x 1 0 x2 0 0:


x1 a 1 sin 2t a 2 cos 2t b 1 sin 4t b 2 cos 4t 14 sin t,
x2 a 1 sin 2t a 2 cos 2t b 1 sin 4t b 2 cos 4t 16 sin t.
0 a2 b2, 0 a2 b2;
a2 b2, b2 b2 ; b2 0, a2 0.
The resulting particular solution from is:
x1 t cos 4t 15 cos 2t 14 cos t,
x2 t cos 4t 15 cos 2t 16 cos t.

"Describe it as a superposition of oscillations at three different frequencies."

We have a superposition of three oscillations, in which the two masses:


Move in opposite directions with frequency 3 4 and equal amplitudes.
Move in the same direction with frequency 2 2 and equal amplitudes;
Move in the same direction with frequency 1 1 and with the amplitude of motion of m 2 being
16, and m 1 being 14.

Problem: #11a Consider a mass-and-spring system containing two masses m 1 1 and m 2 1


whose displacement functions x t and y t satisfy the differential
equations: x 40x 8y, y 12x 60y.
What are the natural frequencies, and in what directions and amplitudes do the masses move?

40 8
A ,
12 60

Determining the eigenvalues:


40 8 2
40 60 96 100 2304
12 60
64 36 . So: 1,2 36, 64.

40 36 8 4 8 4 8
1 36 :
12 60 36 12 24 0 0

1 2 T
, y s, and x 2s, so v 1 21 , where s 1.
0 0

T
Similarly for 2 64 : v 2 1 3 .
2 1
So we have the general solution: x a 1 cos 6t b 1 sin 6t a 2 cos 8t b 2 sin 8t
1 3
OR
xt 2a 1 cos 6t 2b 1 sin 6t a 2 cos 8t b 2 sin 8t,
yt a 1 cos 6t b 1 sin 6t 3a 2 cos 8t 3b 2 sin 8t.
What are the natural frequencies, and in what directions and amplitudes do the masses move?

Problem: #11b Assume that the two masses above start in motion with the initial conditions:
x0 19, x 0 12, and y 0 3, y 0 6, with no external force. Describe the resulting
motion as a superposition of oscillations at two different frequencies.

Applying the first set of initial conditions:


20 2a 1 cos 0 2b 1 sin 0 a 2 cos 0 b 2 sin 0,
3 a 1 cos 0 b 1 sin 0 3a 2 cos 0 3b 2 sin 0.

Simplifying:
20 2a 1 a2, 3 a1 3a 2 .
Solving two equations in two unknowns:
a 1 3 3a 2 , 20 2 3 3a 2 a2 6 7a 2 , a2 2
a1 3 6 9
x t 12a 1 sin 6t 12b 1 cos 6t 8a 2 sin 8t 8b 2 cos 8t,
y t 6a 1 sin 6t 6b 1 cos 6t 24a 2 sin 8t 24b 2 cos 8t.

Applying the derivative initial conditions:


12 12a 1 sin 0 12b 1 cos 0 8a 2 sin 0 8b 2 cos 0,
6 6a 1 sin 0 6b 1 cos 0 24a 2 sin 0 24b 2 cos 0.

Simplifying:
12 12b 1 8b 2 ,
6 6b 1 24b 2 .
Solving two equations in two unknowns:
b 1 1 4b 2 , 12 12 1 4b 2 8b 2 12 40b 2 , b2 0,
b 1 1.
So: xt 18 cos 6t 2 sin 6t 2 cos 8t,
yt 9 cos 6t sin 6t 6 cos 8t.
Describe the resulting motion as a superposition of oscillations at three different frequencies.

1
Problem: #15. Suppose that m 1 2, m 2 2
, k 1 75, k 2 25, F 0 0 100 , and 10
(all in mks units) in the forced mass-and-spring system shown. Find the solution of the system
Mx K x F that satisfies the initial conditions x 0 x 0 0.

Recall: For the spring constants, we have this stiffness matrix:


k1 k2 k2 100 25
K .
k2 k2 25 25
1
2 0 1 2
0
Mass matrix: M 1
, M
0 2 0 2

First we need the general solution of the homogeneous system x M 1 K x , with


1
50 25 2
M K A .
50 50

The eigenvalues of A are 1 25 and 2 75, so the natural frequencies of the system are 1 5
T T
and 2 5 3 . Associated eigenvectors are v 1 1 2 and v 2 1 2 .

So the complementary solution x c t is given by


x1 t a 1 cos 5t b 1 sin 5t a 2 cos 5 3 t b 2 sin 5 3 t ,
x2 t 2a 1 cos 5t 2b 1 sin 5t 2a 2 cos 5 3 t 2b 2 sin 5 3 t .
T
Trial solution to "F 0 0 100 , and 10" is...

T
Recall that: x Ax f M 1K x M 1 F 0 cos t M 1K x 0 200 cos 10t

(note from image above that F 0 is only directly affecting m 2 ).

T
So trial solution: x trial t c1 c2 cos 10t, and we find...
x trial 10 c sin 10t, x trial 100 c cos 10t.

T
x trial A x trial 0 200 cos 10t

Substituting...
25
c1 50 2 c1 0
100 cos 10t cos 10t cos 10t,
c2 50 50 c2 200

25
100c 1 50c 1 2
c2
, (two equations in two unknowns)
100c 2 50c 1 50c 2 200

25 1
50c 1 2
c2, c1 c
4 2
50c 2 50c 1 200 50 14 c 2 200
1 3 16 4
c2 c
4 2
4, c
4 2
4, c2 3
and c 1 3
.

So a particular solution x sp t is described by


4 16
x sp 1 t 3
cos 10t, x sp 2 t 3
cos 10t.
General Solution:
x t xc t x sp t
4
x1 t a 1 cos 5t a 2 sin 5t b 1 cos 5 3 t b 2 sin 5 3 t 3
cos 10t,
8
x2 t 2a 1 cos 5t 2a 2 sin 5t 2b 1 cos 5 3 t 2b 2 sin 5 3 t 3
cos 10t.

" Initial conditions x 0 x 0 0"


Finally, when we impose the initial conditions on the solution x t xc t x sp t
2 4
0 a 1 cos 0 0 b 1 cos 0 0 3
cos 0 a1 b1 3
,
8 16
0 2a 1 cos 0 0 2b 1 cos 0 0 3
cos 0 2a 1 2b 1 3
.

4 4 16
a1 b1 3
, 2b 1 2 b1 3 3
, 4b 1 8,
2
b1 2, a1 3
.

2
We find that a 1 3
, a2 0, b 1 2, and b 2 0.
Thus the solution we seek is described by
2 4
x1 t 3
cos 5t 2 cos 5 3 t 3
cos 10t,
4 16
x2 t 3
cos 5t 4 cos 5 3 t 3
cos 10t.

We have a superposition of 2 natural oscillations with the frequencies 1 5 and 2 5 3 and


forced oscillation with 10. In each of the two natural oscillations the amplitude of motion of m 2
is twice that of m 1 , while in the forced oscillation the amplitude of motion of m 2 is four times that of
m 1 . Regarding direction of motion, in oscillation 5 the masses are moving in the same direction,
while in the other two oscillations they are moving in opposite directions.

You might also like