Chapter 2

Mathematical Modelling
Electrical Systems
R L C
Voltage (e), Current (i)
Mathematical Modelling

iR = e / R
Mathematical Modelling

iR = e / R
iC= C de / dt
Mathematical Modelling

iR = e / R
iC = C de / dt
iL = (1/L) ∫ e dt
Mechanical Systems
M K B
velocity ( v)
force ( f )
f M = M dv/dt
f M = M dv/dt
f K = K ∫ v dt
f M = M dv/dt
f K = K ∫ v dt
f B = Bv
Force - current analogy

iC = C de/dt : fM = M dv/dt

iL = 1/L edt : fK = K ∫ vdt

iR = e/R : fB = Bv
Force - current analogy
C M
Force - current analogy
C M
1/L K
Force - current analogy
C M
1/L K
1/R B
Integro - differential equations

by
● analogyapproach
●free body diagram
approach
Electro mechanical example

K
i R L B
+
eS +

- M
f v
● voltage source eS
● current i
● force f
● velocity v
● back emf eb
 K B
i R L
+ +
eS eb +

- - M
f v
es
es i f
es i f v
es i f v

eb
 (es - eb)
es + i f v

-
eb
 Equations :
(e s-e b ) = Ri + L di/dt
 Equations :
(e s-e b ) = Ri + L di/dt
K1 i = f
 Equations :
(e s-e b ) = Ri + L di/dt
K1 i = f
f = M dv/dt+B v+K ∫ vdt
 Equations :
(e s-e b ) = Ri + L di/dt
K1i = f
f = M dv/dt+Bv+K ∫ vdt
eb = K 2 v
input output
x K y

y = K x
In our example

i K1
f

v eb
K2
How do we represent
(es - eb) = Ri + L di / dt
Let es - eb = e
Then es + e
Error Detector
-
eb
Comparator
Now
e i
?
Take Laplace
Transform of
e = Ri + L di/dt
E (s) = R I (s) + Ls I (s)

= (R + Ls) I(s)

assuming ?
Assuming
Initial Conditions
to be Ignored
E I
?
 E I
?
I(s)/E(s) =
1/(R + Ls) = G1 (s) (say)
Similarly

F v
?
Mechanical Components
+
K
M f
B
Free body diagram approach:
x
Kx f
M
°

Bx
 °° °
M x = f - Kx - B x
°° °
M x + B x + Kx = f
 Analogy approach
x

B M K f

g
 F v
?
? = 1 / ( Ms + B + K /s )

= G2 (say)
es + e i f v
G1 K1 G2
-
eb

K2
Block Diagram Reduction
x1 x2 x3
G1 G2

X2 / X1 = G1
X3 / X2 = G2
Therefore

X3 / X1 = G1 . G2
es + e G1 G 2 K 1 v
-
eb

K2
Let us take

R + E G C
-

where E = R-CH & C = GE

 Eliminating E ,

(R - CH) G = C

C G
or ------ = -----------
R 1 + GH
es + e v
G1 G 2 K 1
-

K2

V (s) G1 G2 K1
--------- = ----------------------------
E s(s) 1+
Block Diagram Reduction

G1 G2 G3

becomes

G1 G2 G3
Block Diagram Reduction

G1
+

+
G2

becomes
G1 + G2
Block Diagram Reduction
z
+
x +
becomes
-
y z
+
x + +
-
y
Block Diagram Reduction

R + C
G
-
H

C G

R 1+ GH
Block Diagram Reduction

R + C
G
+
H

C G

R 1- GH
Block Diagram Reduction
G1 G2

is same as
G1 G2

1/G2
Block Diagram Reduction
G1 G2

is same as
G1 G2

G1
Position
Control
Example
 Components
Load (Inertia & Friction)
DC motor
Reference POT
Feedback POT
Amplifier
 Ra
Motor
Load
eb θL θr
va

KA

- -
+ +
 System
Parameters
KP = sensitivity of the POT’s
in V / rad
KA = amplifier gain in V / V
 System
Parameters
Ra = armature resistance
for dc motor in ohms
KT = Torque constant of dc
motor in Nm / A
System Parameters
JL = Moment of inertia of
the load in Kg.m 2

BL = Coefficient of friction in
Nm /rad / sec
Reference signal / Input/
Command

θr ?

Ref. POT

? = KP θr Volts
Feedback Signal

θL ?

Feedback POT

? = KP θL Volts
 Error Signal
θr KP + e
- θL
KP
f.b. POT

e = θ r K P - θ L KP
= K P ( θr - θ L ) volts
e = KP ( θ r - θ L )

θr + KP
e
-
θL
 Amplifier
θr + KP
e Va
-
θL
 + Amplifier
θr + e Va +
KP KA
- -
θL eb
 Amplifier
θr + e va + dc motor
ia
Kp KA
- -
θL eb
 Amplifier
dc motor
e va + ia T
θr
+
Kp KA 1/Ra
- -
eb
θL
 Amplifier dc motor
e va ia T
θr+ Kp KA + 1/Ra KT

- -e
b
θL
 Amplifier dc motor Load
+ e va ia T
θr Kp KA 1/Ra KT 1/(J s+B )s θL
- + L L

-
θL
eb
 Amplifier dc motor Load
+ e va 1 ia θL
θr KP KA + KT T 1
- Ra (JLs+BL)s

- eb
K b ωL s
θL
Block Digram Reduction

θr + va + KT / Ra θL
KPKA (JLs+BL)s
_ _
s Kb
θL

Using G1 G2 ≡ G1 G2
Now use

R + C
G
_
H

≡ R G
1 + GH
C
 K T / RA
θr + (JLs+BL)s θL
KP K A
s Kb KT/Ra
1+
- (JLs+BL)s

θL
1
 θ L
(s)
/ θ (s)
r
KP KA KT / Ra
=
JLs2 + s ( BL + Kb KT / Ra) + KP KA KT / Ra
 Control by
1. Armature voltage (va)
2. Field current (if)
We know
Motor Torque
TM is proportional
to the product of flux and
armature current
i. e., TM = Kφia
Assume
Field current as constant;

Then flux is also constant.

Make if constant,

Then TM = KT ia
KT in Nm / A

.... armature Controlled

dc servo motor
Now
about ....

field controlled

dc motor....
Assuming
The magnetic circuit is linear

i. e. Flux = K if
and TM = K if ia
The torque equation is still
non linear
 Make ia Constant

Then TM = K T i f
KT in Nm/ field amperes
- field Controlled
dc servo motor
Field Controlled dc servo
motor
Ia = Constant
Rf , L f
θL θr
JL , BL
if _
vf KA E
+
 Equations

E = KP ( θ r - θ L)

Vf = KAE
If = Vf / (R f + sL f )
TM = K T I f
θL = TM /(JLs2 + s BL )
θr + vf 1 if 1 θL
KP KA ----------- KT -----------
- Rf + s Lf JLs2 + BLs
Motor - Load
Direct Coupling

TM TL J L ,B L

T M = TL
ωM = ωL
 Motor - Load
Direct Coupling

TM = TL 1 ωL 1/s θL
JLs + BL
Motor - Load
Gear Reduction
N1

TM TL , ωL
ωM

JL, BL

N2
TM
=
N1
=
ωL
TL N2
ωM
TM TL
N2 / N1
 TM
=
N1
=
ωL
TL N2
ωM
TM TL 1/(J s+B )
N2 / N1 ωL θL
L L
1/s
 TM
=
N1
=
ωL
TL N2
ωM
TL
TM
N2 / N 1
1
------------ ωL θL
J Ls + B L 1/s

ωM
N2/N1 ωL s
 Suppose Motor also has
JM, BM

N1

JM , BM T2
T1

JL , BL
N2
TM =

(JM s +BM) ωM +T1 ( say )

 ωL =
T1
=
N1

ωM T2 N2

T2 = ( JLs + BL ) ω
L
Eliminate T 1, T 2, ωL,TL
TM = (JM s + BM) ωM +

(N1/N2) (JLs + BL) (N1/N2) ωM

 Let (N1 / N2) = a

TM = {( JM + a2 JL )s + (BM + a2 BL)} ωM
Signal flow graph
x Y
G

x G Y

Y = XG
x1 1 X3

2
x2

x3 = x1 + 2x2
x1 1 X3 -1
x4
2 1
x2 x5

x4 = -x3 + x5
Let us take

R + E C
G
-

where E = R-CH & C = GE

Equations E = R - CH
C = GE

Variables R
E
C
R 1 E G C 1 ?

-H

? = C
Mason’s gain formula

Forward path gain = G

Loop gain = - GH
 Forward path gain
C/R = ------------------------
1- loop gain

G
= -------------------
1 - ( - GH )
 Equations

Block diagram Signal flow graph

Reduction Gain formula

Transfer Function
 G2
+
R + E + C
G1
-
H

E = R - CH
C = EG1 + EG2
 E = R - CH
C = EG1 + EG2

G2
R 1 E G1 C 1 C

-H
 G1 + G 2
C/R = -----------------------
1 + G1 H + G2H
 G2
E X + C
R + G1
+
-
H
E=R-XH
X = EG1
C = X + E G2
E = R - XH
X = EG1
C = X + EG2
G2

R 1 E G1 X 1 C 1 C

-H
 G1 + G2
C / R = --------------
1 + G1 H
C R =
∑ PK ∆ K

∆
R 1 X1 G1 X2 G2 X3 G3 C 1 C

- H1 - H2 - H3

Forward path gain

P1 = G1 G2 G3
R 1 X1 G1 X2 G2 X3 G3 C 1 C

- H1 - H2 - H3

Loops: - G1H1
- G2H2
- G3H3
R 1 X1 G1 X2 G2 X3 G3 C 1 C

- H1 - H2 - H3

∑( loop gains) =
- G1H1 - G2H2 - G3H3
R 1 X1 G1 X2 G2 X3 G3 C 1 C

- H1 - H2 - H3

Out of 3 loops,
2 loops do not touch each other;
X1 G1 X2 X3 G3 C

- H1 - H3
Their gain product

= (- G1H1) ( - G3 H3 )
∆ = 1 - ∑ ( loop gains)
+ ∑ gain products
of non touching loops
two taken at a time
- etc.
 =

1 - ( - G 1 H1 - G 2 H2 - G 3 H3 )
+ (-G1H1 ) (-G3H3 )
Loop not touching
the forward path ?
Nil.
∴ ∆1 = 1
 P1 ∆
C/R = 1

∆
( G1 G2 G3 ) (1)
=
1 + G1 H1 + G2 H2 + G3 H3 + G1 H1 G3 H3
 G2

R E X + C
+ G1
+
-
H
E = R-XH
X = EG1
C = RG + X
E = R - XH
X = EG1
C = RG2 + X
R 1 E G1 X 1 C 1 C

-H
G2
 G 1 + G 2 ( 1 + G1 H )
C/R = --------------------------
1 + G1 H
Multi input multi output
R1 + E1 C1
G1
-
) H1

_ H2
R2 + E2 C2
G2
To find C1 / R1 ,

assume R2 = 0
R1 + E1 C1
G1
-
H1

-1 G2

H2
C1/ R1 =
G1
1 + ( - G1 G2 H1 H2 )
By signal flow graph.
R1 1 E1 G1 C1 1 C1

-H2 -H1

R2 1 E2 G2 C2 1 C2
C1 / R1
forward path gain = G1
loop gain = G1 G2 H1 H2
G1
C1 / R1 =
1 - G1 G2 H1 H2
C2 / R1
forward path gain

= - G1 G2 H1

loop gain = G1 G2 H1 H2
 - G1 G2 H1
C2 / R1 =
1 - G1 G2 H1 H2
 Ra
Motor
Load θL θr
va eb

KA

- -
+ +
 E = KP ( θr - θL)
Va = KA E
TM = ( KT/Ra) (Va - Eb)
θL/ T =1/ (J s2 + B s ) = G (say)
M L L 1

Eb / θL = sKb = H1 (Say)
Variables : θr
E
Va 5 equations
TM needed
Eb
θ
θr KP E KA Va KT/Ra TM G1 θL 1 θL
- KT/Ra

1
H
Eb

- KP
forward path gain = KP KA KT G1 / Ra
Loop gains - G1 H1 KT / Ra
- KA KT G1 KP/ Ra
 forward path gain
θ L/θ r = --------------------------
1 - ∑ ( loop gain )
θL/θr
KP KA KT G1/Ra
= ----------------------------------------------------
1- (-G1H1K T / Ra) - ( - K A K T G1KP/Ra)
 Disturbance input
T M - T D = TL
TD
- 1
TM + TL ------------ θL
JLs2 +BLs
Liquid level control
 Components
: Tank with inflow valve
: float & f. b. POT.
: ref POT
 Components
: d.c. amplifier
: armature controlled
dc. servo motor
: valve at inlet
Tank
Q in

H Q out

A , area of C. S.
Q in - Q out = d/dt (AH)
and Q out = H /R
allow small variation in
Q in
Q in → Q in + ∆ Q in

∴ H → H+∆ H

Q out → Q out + ∆ Q out

Q in + ∆ Q - (Q out + ∆ Q out )
in
d
= ---- A (H + ∆ H )
dt

Q out = ( H + ∆ H ) /R
 d
∆ Q in - ∆ Q out = ---- A ∆ H
dt

∆ Q out = ∆ H / R

Let A = C , capacity of tank

Let A = C , capacity of tank

Taking Laplace transforms

∆ Q in (s) - ∆ Q out (s)

= Cs ∆ H (s)

∆ Q out (s) = ∆ H/R

with ∆ Q in as input and ∆ H
as output

∆ H (s)/∆ Q in (s) = R / (1+RCs)

RC , time Constant
S im ila r ly

θ r in p u t P O T a n g le

θr → θr+ ∆ θr

K P P O T g a in in V / r a d
θL f. b . P O T a n g le

θL → θL + ∆ θL

K P P O T g a in in V /r a d
Float linkage

∆ H × Kf = ∆ θ L

Kf rad / m
Amplifier gain KA V / V

dc motor armature
Ra Ohms
La Henries
KT Nm / A
K v / rad / sec
Valve Jv , Bv
KV Valve Orifice Coefficient
θV , Valve opening
θv θv + ∆ θv
Variables
∆ θr ∆ TM
∆ θL ∆ θv
∆ e ∆ Q in
∆ va ∆ H
∆ eb
∆ ia
In p u t ∆ θr

O u tp u t ∆ H
 Block diagram
∆ia ∆TM
∆θ ∆va ∆θ
r 1
+ + 1 1 v
KP K KT
-
A
Ra+s La JVs + BV s
-

∆θ ∆e b
L Kb
∆θ ∆ Q in R ∆H
v Kv ------
1+RCS

∆θ
L Kf
Liquid level Control
We had taken
valve out flow as
Qin = Kv θ v
Suppose it depends on
pressure of supply also,
then

Q in = K v θ v+ K r p
In incremental form

∆Q = Kv ∆ θ + K ∆p
in v r
Modify the block diagram as

∆p

θv Kr

+ ∆ Q in
∆ TM 1 ∆θ v +
---------- Kv
JVs2 + BVs
 Find transfer
functions
∆ H (s)/ ∆ θ (s) with ∆ p = 0
r

∆ H (s) / ∆ p (s) with ∆ θr= 0

 Temperature Control

Ti
+
e vh
T
T -

T
Components
plant
Transducer gain K t V/ deg

reference source vr V
error e = vr - Kt T
in volts
vr = ref. voltage input

T = temperature to be
controlled
amplifier gain KA V / V
heating element
vh = e KA
vh = voltage input to
heater
KA = amplifier gain
in V / V
 2
vh
H= ---------- cal
RJ

H = heat input into plant

R = resistance of heater
 Plant
Heat input thro' input stream
= Q c Ti cal
Q = mass in flow
c = specific heat
Ti = inlet temperature
Heat outflow thro'
output stream
= Q c T cal
Therefore
Q c Ti - Q c T + H
= McsT
M = mass hold up of
tank
Ti - T
+ H = ( Mc ) s T
1/ Q c

Ti - T
+ H = C s T
R
Where

R = 1/Qc

C = Mc
Incremental Relations
Let
vr → vr + ∆ vr

then
e → e+∆ e
vh → vh + ∆ vh
H → H + ∆ H

T → T + ∆ T
 e = vr - Kt T

e + ∆ e = (vr +∆ vr) - Kt
(T +∆ T)

∴ ∆ e = ∆ vr - Kt ∆ T
 vh = e K A

vh + ∆ vh = (e+∆ e)KA

∴ ∆ vh = K A ∆ e
 2
H = vh / RJ
2
H +∆ H = (vh +∆ vh ) / RJ
2 2
(vh +∆ vh) - vh
∴ ∆ H = ---------------------
RJ
 2 vh ∆ vh
= -------------
RJ

= Kh ∆ vh
 Ti- T
------- + H = Cs T
R

T i - (T + ∆ T )
---------------- + H + ∆ H = C s (T + ∆ T )
R
 -∆T
∴ ----- + ∆ H = Cs∆ T
R

or ∆ T / ∆ H

= R / (1 + RCs)
∆e = ∆ vr - Kt ∆ T

∆ vh = KA ∆ e

∆ H = Kh ∆ vh

∆T = ∆ H {R / (1 + RCs)}
= ∆ H G (say)
∆ r 1
v ∆e KA ∆ v h Kh ∆H G ∆T 1 ∆T

-Kt
forward path gain = KA Kh G

loop gain = - KA Kh G Kt

∆ T KA Kh G
------- = ------------------------
∆ vr ( 1 + KA Kh G Kt )
Suppose input temperature
Ti also varies
Ti → Ti + ∆ Ti

then Ti - T
--------- + H = CsT
R
Ti + ∆ Ti - (T+∆ T)
------------------------ + H + ∆ H
R
= Cs ( T+ ∆ T )

∆ Ti - ∆ T
------------- + ∆ H = Cs ∆ T
R