Chapter 3

Mathematical Modeling of
Mechanical Systems and Electrical
Systems
 3-1 Introduction

This chapter presents mathematical modeling of

mechanical systems and electrical systems.
The mathematical law governing mechanical
systems is Newton’s second law, while the basic
laws governing the electrical circuits are
Kirchhoff’s laws.
3-2 Mathematical modeling of mechanical systems

Typical mechanical systems may involve two kinds of

motion: linear motion and rotational motion. Spring,
mass, damper and inverted pendulum are widely used
devices to describe a large class of mechanical systems.

Example (Automobile suspension system). As the car

moves along the road, the vertical displacements at the tires
act as the motion excitation to the automobile suspension
system. The motion of this system consists of
a translational motion of the center of mass and a rotational
motion about the center of mass, and can be simplified as a
system with springs, mass and dampers shown below.
3-2 Mathematical modeling of mechanical systems
Example. Equivalent spring constants.
k1 x
k1x  k 2x  F  keq  k1  k 2
F
k1y  F , k 2 (x  y )  F
k2 F
 k 2 (x  )  F
y x k1
F 1
keq  
F x 1 1

k1 k2 k1 k 2
Systems consisting of two springs in parallel and
series, respectively.
Example. Systems consisting of two dampers connected
in parallel and series, respectively. Find their equivalent
viscous-friction coefficients beq with respect to dy/dt and
dx/dt.
b1

b2
(Fig.a)
x y

Solution: (a) The force f that causes the displacements

is
f  b1 (y  x )  b2 (y  x )
 (b1  b2 )(y  x )
 beq (y  x )
(b): b1 b2

(Fig.b)
x z y

The same force f is transmitted through the shaft and

therefore,
f  b1 (z  x )  b2 (y  z ) (1)

from which we have

1
(b1  b2 )z  b2y  b1x  z  (b2y  b1x ) (2)
(b1  b2 )
Since we want to determine the relationship between dy/dt
and dx/dt, that is,
f  beq (y  x )
where beq is the equivalent viscous friction coefficient, by
substituting (2) into (1) yields.
 1 
f  b2  y  (b2y  b1x ) 
 (b1  b2 ) 
bb
 1 2 (y  x )
(b1  b2 )

bb 1
Hence, beq  1 2

(b1  b2 ) 1  1
b1 b2
Example. Consider the spring-mass-dashpot system
mounted on a massless cart. Let u, the displacement of the
cart, be the input, and y, the displacement of the mass, be
the output. Obtain the mathematical model of the SMD
system.

u y u y

Forces acting on the

mass:
Massless cart k

m
 du dy 
b   k  u y  b
 dt dt 
By Newton’s law: d 2y  du dy 
m b   k  u y 
dt  dt dt 
d 2y dy du
m  b  ky  b  ku
dt dt dt
Therefore, the system transfer function can be obtained
as u y

Y (s ) bs  k

U (s ) ms 2  bs  k
Massless cart k

b
Example. Mechanical system is shown below. Obtain the
transfer functions X1(s)/U(s) and X2(s)/U(s), where u denotes
the force, and x1 and x2 denote the displacements of the two
carts, respectively.
Solution: By Newton’s second law, we have
m 1x1  u  k1x 1  k 2 (x 1  x 2 )  b(x1  x 2 )
m 2x2  k 3x 2  k 2 (x 2  x 1 )  b(x 2  x1 )

u x1 x2
k2

m1 m2
k1 k3
b
Simplifying, we obtain

m 1x1  bx1  (k1  k 2 )x 1  u  bx 2  k 2x 2

m 2x2  bx 2  (k 2  k 3 )x 2  bx 1  k 2x 1

u x1 x2
k2

m1 m2
k1 k3
b
Taking the Laplace transforms of these two equations, and
assuming zero initial conditions,
[m 1s 2  bs  (k1  k 2 )]X 1 (s )  U  (bs  k 2 )X 2 (s ) (1)

[m 2s 2  bs  (k 2  k 3 )]X 2 (s )  (bs  k 2 )X 1 (s ) (2)

Solving Equation (2) for X2(s) and substituting it into

Equation (1) and simplifying, we get

X 1 (s )

U (s )
[m 2s 2  bs  (k 2  k 3 )]
[m 1s 2  bs  (k1  k 2 )][m 2s 2  bs  (k 2  k 3 )]  (bs  k 2 ) 2
Example. An inverted pendulum mounted on a motor-
drive cart is shown below. This is a model of attitude
control of a space booster on takeoff; that
is, the objective of the y

attitude control problem is

to keep the space booster x


in a vertical position.  sin 

l cos 
mg
u  

O
x
M
u
Define:
u: the control force to the cart;
: the angle of the rod from the vertical line;
(xG, yG): the center of gravity of
y
the rod in (x, y)
coordinates. Hence,
xG  x  l sin q 

x 
yG  l cos q

1) The horizontal V

motion of cart is mg

described by H H
O x

d 2x u V
M
M 2  u H (1)
dt
 2)The horizontal motion of center of gravity of the rod is

2
d
m 2 (x  l sin q)  H (2)
dt y


3)The vertical motion of 

center of gravity of the x

pendulum rod is

2 V
d
m 2 (l cos q) V  mg (3) mg
dt
H H
O x

M
u V
4) The rotation motion of the rod about its center of
gravity can be described by
I q  l (V sin q)  l (H cos q) (4)
y

where I is the moment of  

inertia of the rod about x
its center of gravity.

5) Linearization of the V
equations (2)-(4): mg

H H
O x

M
u V
Since we must keep the inverted pendulum vertical, we can
assume that  and d/dt are small quantities so that sin,
cos1 and (d/dt)2 0. Hence, the equations (2)-(4) can be
linearized as
m (x  l q)  H (2)

0 V  mg (3)  Mx  u  H (1)
 
I q V ql  Hl (4)
By replacing (1) into (2) yields
(M  m )x  ml q  u (5)
From equations (2)-(4), we have
I q  mgl q  Hl  mgl q  l (mx  ml q)
 (I  ml 2 )q  lmx  mgl q (6)
 From (6), we have
mgl q  (I  ml )q
2 
x  (7)
lm
Substituting (7) into (5) yields
mgl q  (I  ml 2 )q
(M  m )  ml q  u (8)
lm
Taking the Laplace transform on both sides of (8) yields
mgl   (I  ml 2 )s 2 
(M  m )  mls 2   U (s ) (9)
lm
Therefore,
(s ) lm

U (s ) (M  m )mgl  (MI  mMl 2  mI )s 2 
 y
Example. An inverted x  sin

pendulum is shown in the

figure below. Similar to the mg
previous example:
cos 


u 
O x
P
M
u

(s ) lm

U (s ) (M  m )mgl  (MI  mMl 2  mI )s 2 
For this case, the moment of inertia of the pendulum
about its center of gravity is small, and we assume I = 0.
Then the mathematical model for this system becomes:

(s ) 1

U (s ) (M  m )g  Mls 2 
Example. Gear Train: Rotational transformer. In
servo control systems, the rotor drives a load through
gear train.

Jm N1
m m L, L
N2 JL

Gear ratio (transmission ratio):

N 1 qL wL
n  
N 2 qm wm
where N1 and N2 denote the number of teeth of the two
gears, respectively.
3-3 Mathematical modeling of electrical systems
1. RLC circuit
Example. RLC circuit
u y
RLC

By using Kirchihoff’s law, we have

u  iR  y  v L
Since
dy di d 2y
i C , v L  L  LC 2
dt dt dt
we have
d 2y dy
CL 2  CR y  u
dt dt
 Example. RLC circuit. Find V(s)/R(s).

r(t)
Current v(t)
source

v (t ) dv (t ) 1 t

R
C
dt

L  v (t )dt  r (t )
0

Taking the Laplace transform of the above equation yields

V (s ) 1
 CsV (s )  V (s )  R (s )
R Ls
Hence,

1 1
V (s )(  Cs  )  R (s )
R Ls
V (s ) 1
 
R (s ) ( 1  Cs  1 )
R Ls
R Ls

R LCs 2  Ls  R
 2. Transfer functions of cascaded elements

Example. Find Eo(s)/Ei(s).

Note that in this circuit, the second portion (R2C2)

produces a loading effect on the first stage (R1C1 portion);
that is , we cannot obtain the transfer function as we did
for transfer functions in cascade.
 ei  i1R 1  uc1

 duc1 1
C 1 dt  i1  i 2  uc1  C  (i1  i2 )dt
 1
and
 1
 o 2 2 C  (i1  i 2 )dt
e  i R 
  1    
 uc 1

 1
eo 
 C2  i 2dt

Taking the Laplace transforms of the above equations,

we obtain
1
I1  [E i U c1 ]
R1
 1
U c1  [I 1 (s )  I 2 (s )]
C 1s
1
I 2 (s )  [U c1  E o (s )]
R2
1
E 0 (s )  I 2 (s )
C 2s

Ei 1 I1 1 U c1 1
I2 1 Eo
R1 sC 1 R2 sC 2
U c1

from which we obtain

E 0 (s ) 1

E i (s ) R 1C 1R 2C 2s 2  (R1C 1  R 2C 2  R1C 2 )s  1
The above analysis shows that, if two RC circuits
connected in cascade so that the output from the first
circuit is the input to the second, the overall transfer
function is not the product of 1/(R1C1s+1) and
1/(R2C2s+1) due to the loading effect (a certain amount
of power is withdrawn).
 3. Complex impedances

Resistance R: R
Capacitance: 1/Cs
Inductance: Ls
Example. Find Y(s)/U(s)

Solution:
Y (s ) 1 / Cs

U (s ) R  Ls  1 / Cs
 Example. Find Eo(s)/Ei(s).

Solution: Utilizing complex impedance approach, we

have (from left to right)
  E i (s ) U ab (s ) 1
I 1 (s )  R1

R1
[E i (s ) U ab (s )]

 1
U
 ab (s )  [I 1 (s )  I 2 (s )]
 sC 1

I (s )  1 [U (s )  E (s )]
 2 R2
ab 0


E (s )  1 I (s )


0
sC 2
2

Ei 1 I1 1 U ab I2 1 Eo
1
R1 sC 1 R2 sC 2
U ab
Example. Find Eo(s)/Ei(s).

C
4. Transfer functions of nonloading cascaded
elements
Again, consider the two simple RC circuits. Now, the
circuits are isolated by an amplifier as shown below and
therefore, have negligible loading effects, and the
transfer function
E 0 (s ) K

E i (s ) (R 1C 1s  1)(R 2C 2s  1)

R1 a R2
+ +
Isolating
amplifier
ei i1 C1 (Gain K ) i2 C2 eo

- -
b
In general, the transfer function of a system consisting
of two or more nonloading cascaded elements can be
obtained by eliminating the intermediate inputs and
outputs:

R(s) U C(s)
G1(s) G2(s)

R(s) C(s)
G1(s)G2(s)

C (s )  G 2 (s )U (s )  G 2 (s )G 1 (s )R (s )
5. Operational amplifiers

In the ideal op amp, no current flows into the input

terminals, and the output voltage is not affected by the
load connected to the output terminal. In other words,
the input impedance is infinity and the output impedance
is zero.
6. Examples: DC motor and servo system

Example. Find the transfer function of a DC motor.

Assume that the rotor has inertia Jm and viscous friction
coefficient fm. Determine m(s)/Ua(s), where m is the
rotational angle and ua is the power supply voltage
.

Ua(s) DC m(s)
motor
 Stator winding: 定子绕
组；
Rotor windings: 转子绕
组；
Shaft: 电机轴；
Inertia load: 惯性负载 ;
Bearings: 轴承
Commutator ：换向器
ua J m

fm

m
Rotor: 转子； Torque: 电磁转矩； Back emf voltage:
反电势； Armature current: 电枢电流 ;

ua Jm
fm

m

m: rotational angle; ua: power supply voltage

The motor equations:
1) The electromagnetic torque Mm on the rotor in terms
of the armature current ia (Cm: torque constant):
M m  C m ia (1)
2) The back emf voltage Eb in terms of the shaft’s
rotational velocity dm/dt is (Kb : electrical constant):

E b  K b qm (2)
Ra 3) The electrical equation:
J m , fm
dia
ua
ia
La ua  ia R a  La  Eb (3)
dt
qm , wm
4) The torque equation: by using Newton’s law,
J m qm  M m  fm qm  J m qm  fm qm  M m (4)

Taking the Laplace transform of (3) and in view of (2)

yields
U a  I a R a  LasI a  K bs m
Hence,
1
Ia  [U a  K bs m ] (5)
Las  R a
Taking the Laplace transform of (4) and in view of (1)
yields
1
m  CmIa (6)
J m s  fm s
2
 From (5) and (6), we have
ua 1 Cm m
Las  Ra s (J m s  fm )

K bs
Eb

m Cm

U a (s ) [(Las  R a )(J m s 2  fm s )  C m K bs ]

In particular, if La0, we can simplify the model as

m Km

where U a (s ) s (T m s  1)
Cm J m Ra
Km  Tm 
R a fm  C m K b R a fm  C m K b
Example. A servo system is shown below, where
r and c are the input and output (that are propor-tional to
the angular positions of potentiometers),
e=rc, (erec)=K0(rc)
K0: proportionality constant; K1: amplifier gain; eb: back
emf voltage; n: gear ratio.
 ia T 
r e ea 1 1 c
1
 K0 K1  La s  R a
K2
J 0s  b0
n
  s
ev
eb
K3

where in this example, for the DC motor

1) The electromagnetic torque (K2: torque constant):
T  K 2ia (1)

2) The back emf voltage (K3 : electrical constant):

eb  K 3q (2)

where  is the rotational angle;

3) The electrical equation:
dia
ea  ia R a  La  eb (3)
dt
4) The torque equation:

J 0q  b0q  T  K 2ia (4)

where J0 and b0 denote the moment of inertia and

friction constant, respectively.
Then, by neglecting La, the open-loop transfer
function can be simplified as
C (s ) K
=
E (s ) s (Js +B)
 r e K c
s (Js  B )

where

J  J 0 / n 2

B  [b 0  (K K
2 3 / R a )] / n 2

K  K K K / R n
 0 1 2 a
 Summary of Chapter 3
In this chapter, we studied the mathematical
modeling of some simple mechanical and electrical
systems.
1. For mechanical systems, Newton’s second law is
applied to establish the mathematical models of (1)
some spring-mass-damper systems and (2), inverted
pendulum systems. y

 

u y x


V
mg
Massless cart k
H H
m O x

b u
M
V
2. For electrical systems, Kirchhoff’s laws are used. In
particular, the concept of complex impedance is
introduced, which provides a convenient way to
obtain system transfer function.
3. The mathematical modeling of DC motor, which
combines mechanical and electrical systems, is
introduced and can be expressed as
M m  C m ia

E
 b  K 
q
b m
m Cm
 dia  
u  i
 a a a a dtR  L  E b
U a (s ) [(La s  R a )(J m s 2
 fm s )  C m K bs ]

J m qm  fm qm  M m
Ra
J m , fm
ia Ua(s) m(s)
ua La DC
qm , wm motor