Instrumentation and Control Assignment

Matt Deñelle Larita Binaguiohan

Hya Shane Tadlas Tejume
BSEE 4-9
April 29, 2024

1 Problem
Consider a voltage divider. The output is v2 measured at R1 , and the input is v1 .
a) Model an equation for v2 as a function of v1 , R1 , and R2 . That is, write an equation for
v2 which yields an open-loop system.
Answer:
In the given circuit by using the Kirchoff’s voltage law the first equation is:

+ V2 -

R1
+ I R2
−

−V1 + IR1 + IR2 = 0

V1 = IR1 + IR2

These equations represent the sum of voltages in a loop with a voltage source V , resistor
R1 , and the voltage drop across R2 represented by R2 and V1 , with the current through
the loop represented by I.
Solving for I:

V2
I=
R1 + R2
V2 = IR1

Substituting I into V2 , we get:

IR1
I=
R1 + R2
R1
I=
R1 + R2

1
 So, the expression for current I in terms of R1 and R2 is:

R1
I=
R1 + R2

b) Model an equation for v2 as a function of v1 , v2 , R1 , and R2 that yields a closed-loop

system.
Answer:
By continuing the solution flow in part a, our second equation is:

−V1 + V2 + V3 = 0
V1 = V2 + V3
V1 = V2 + IRL

This represents the sum of voltages around the loop with V1 , V2 , and V3 as voltage drops.
VD represents the voltage drop across diodes. I represents the current through the resistor
R2 , given by:
V1 − V2
I=
R2
 
V1 − V2
VR = IR1 = R1
R2
V1 R1 − V2 R2
=
R2
V1 R1 V2 R2
V2 = −
R2 R2

These equations describe the relationship between the voltages and currents in the given
circuit, considering the resistors R1 and R2 and a diode VD .

c) Contrast the response curve of output v2 between the two scenarios. Plot the response
curve with your selected values for v1 , R1 , and R2 .
MATLAB code:

2
Response curve result:

3
2 Problem
The task is to design a sensor mechanism to ensure safety in a punching device.
the senario is explicitly describe: ”It must be possible to operate the punching device
from three sides. A notch is to punched in the workpiece by a pneumatic cylinder. The
workpiece may be fed from any one of three work stations. The punching operation is
to be triggered when 2 sensors emit a signal. The optical signaling elements a, b and c
are installed for sensing. The cylinder returns to its retracted position when the part is
pulled out of the device.”

a) Create a flowchart describing the step-by-step procedure required in the task.

Answer:

Figure 1: Fluidsim representation

4
 b) Design and implement the circuit in FluidSIM.
Answer:

Figure 2: Fluidsim representation

c) How are the operational condition achieved in the designed circuit.

Answer:
To achieve the described operational conditions in the designed circuit, you would likely
need a combination of sensors, pneumatic control systems, and logic circuits. Here’s a
rough outline of how it might work:
Sensors Placement: Sensors a, b, and c are positioned strategically to detect the presence
of the workpiece at each of the three work stations. These sensors emit a signal when the
workpiece is present.
Signal Integration: The signals from sensors a and b (or any other combination as re-
quired) are integrated to trigger the punching operation. This integration ensures that
the punching only occurs when the workpiece is present at the selected station.
Pneumatic Control: A pneumatic cylinder is responsible for the punching operation.
When the integrated signal from the sensors indicates that the workpiece is present and the
punching operation should proceed, the pneumatic control system activates the cylinder
to punch the notch in the workpiece.
Cylinder Retraction: After the punching operation is completed, the pneumatic control
system retracts the cylinder to its initial position. This retraction is likely automated to
occur after a set time delay or when the workpiece is removed from the device.

5
Safety Measures: Safety mechanisms should be in place to prevent accidental operation,
such as ensuring that the punching operation can only occur when all necessary conditions
are met (e.g., workpiece present, sensors aligned, etc.).
User Interface (Optional): Depending on the complexity of the system, a user interface
might be included to allow operators to select the desired work station or monitor the
status of the device.
Overall, the circuit would need to integrate sensor inputs, control the pneumatic cylinder,
and ensure safe and reliable operation of the punching device from any of the three sides.
Testing and fine-tuning would likely be necessary to ensure proper functionality.

3 Problem
The task is to design a automated washing system for the scenario describe below. ”Metal
components are to be cleaned in an acid bath. These components are placed in a basket
which is hung from a hook on a double-acting cylinder. A single actuation of a pushbutton
causes the basket to be continuously lowered into and pulled out of the acid bath. The
sequence stops automatically after 10 cycles. Sensors B0 and B1 check the fully retracted
and extended positions, respectively, of the cylinder.”

a Design and implement the circuit in FluidSIM. Answer:

Figure 3: Fluidsim representation

b How is the system able to achieve continuous operation and stop after 10 cycles?
Answer:

6
 Continuous operation is achieved by the cyclic nature of the control logic. Each time
the pushbutton is pressed, the system goes through a complete cycle of lowering the
basket into the acid bath, washing the components, and retrieving the basket. This
cycle repeats until the cycle count reaches 10.
The cycle count is incremented each time the cylinder fully extends and retracts,
as detected by sensors B0 and B1. Once the cycle count reaches the predetermined
limit of 10, the system automatically stops the washing sequence, ensuring that only
10 cycles are completed.
By integrating the control logic with the sensing of cylinder positions and the push-
button input, the system achieves both continuous operation and the ability to stop
after a specific number of cycles.

4 Problem
The response of a controller can be affected greatly by nonlinear measurement data.
Consider a PID controller with gains of (kp = 2, ki = 0.5 and kd = 5), used to control
temperature in the range of 20 to 100 degree Celsius. Signal conditioning has provided a
current that varies from 4 to 20 mA for that temperature range, but it varies nonlinearly
by the relation:
T 2
I = 1.91(1 + )
44.72
The controller reacts to the current, not the actual temperature. The next two problems
study the reaction of the controller to a uniform temperature increase of 5°C over 10s but
for two different setpoints: (1) with a setpoint of 30°C, and (2) with a setpoint of 80°C.

(a) Plot the current versus temperature and note the nonlinearity. In both cases, the
temperature changes by 5°C, which is 6.25
(b) How much does the current change in the two cases? What is this in percent of
current range?
(c) For the two cases, express the error in current as percent of range using equations
of the form ep (t) = kp t; assume the current changes linearly over the 10 s.
(d) Set up the PID controller equations for the two cases of Problem S1 and solve to
obtain an equation for the controller output as a function of time. Use the 2% error-
time equations. Assume the initial controller output is 30% in the first case and 50%
in the second case. Plot these functions to compare response. What is the change
in controller output over 10s in the two cases?

5


The five-degree temperature change is a percent change of 100−20 × 100% = 6.25%,
and this occurs in 10 seconds. Therefore, the error equation for temperature is
ep (t) = [6.25%/10 s]t = 0.625% and it doesn’t matter if it is a change from 30 to
35◦ C or from 80 to 85◦ C. But as noted in the problem statement, the controller
reacts to current.
a) The following plot of current vs. temperature clearly shows the non-linear re-
sponse.
The current changes much more for temperature changes in the high range compared
to the low.

7
 Figure 4: Enter Caption

b) We express the current change by finding the currents at temperatures of 30◦ ,

35◦ , 80◦ , and 85◦ C. Given
 2
T
I = 1.91 1 +
44.72
 2
30
I30 = 1.91 1 + = 5.33 mA
44.72
 2
35
I35 = 1.91 1 + = 6.07 mA
44.72
 2
80
I80 = 1.91 1 + = 14.86 mA
44.72
 2
85
I85 = 1.91 1 + = 16.07 mA
44.72
So in the first case, the current change is

∆I30 = 0.74 mA

and in the second, ∆I80 = 1.21 mA. ⇒ The range of current is, of course, ∆I =
20 − 4 mA = 16 mA so we find the change of current in the two cases as percent of
range as,
0.74
∆P30 = × 100% = 4.63%
16
1.21
∆P80 = × 100% = 7.56%
16
Therefore, even though the temperature changed by the same amount in both cases,
the percent of range change of current was not the same. Thus, the controller reacts

8
differently in the two cases, suggesting the temperature changed more in the second
case. Since the change occurred in 10 seconds, equations for the two cases are found
as,

ep30 (t) = [4.63%/10 s]t = 0.463t%

eP80 (t) = [7.56%/10 s]t = 0.756t%

Using the given gains and ep (t) = kt, the general form of the two equations is
Z t
d
P (t) = 2kt + k tdt + 10 (kt) + P (0)
0 dt
2
kt
= 2kt + + 10k + P (0)
2
We have
t2
 
P30 (t) = 0.463 + 2t + 10 + 30
2
 2 
t
P80 (t) = 0.756 + 2t + 10 + 50
2

The following plot shows the response as a function of time.

After 10 seconds, the changes are:

∆P30 (10) = 37%

∆P80 (10) = 60%

Actually, in the second case, the output saturates at 100% as the plot shows.

9
 We see that the form of the controller response is the same in both cases but in the
80◦ C case, the change is larger in magnitude. Now we set ep (t) = sin(2πf t). The
equation for controller output is found by recalling that,
Z
1
sin(ωt)dt = − cos(ωt)
ω
d
[sin(ωt)] = ω cos(ωt)
dt
1
We have P (t) = sin(2πf t) + 2πf
[− cos(2πf t)]t0
 
1 1
P (t) = sin(2πf t) + 2πf − cos(2πf t) +
2πf 2πf

Using the identities given in the problem, we write this equation in the form,

P (t) = k sin(2πf t + ϕ)

where s  2
1
k = 1 + 2πf −
2πf
 
−1 1
ϕ = tan 2πf −
2πf

5 Problem
Model the transfer function in time domain, frequency domain, and state space for series and
parallel RLC. for parallel RLC, the transfer function concerns current, and for series, it concern
voltage. Provide the block diagram, phase variable form, controller canonical form, and observer
canonical form (including sidnal flow graphs).
Answer:

Figure 5: Parallel RLC Circuit

10
 Figure 6: Series RLC Circuit

6 Frequency Domain
6.1 Parallel RLC Circuit

Figure 7: Parallel RLC in Frequency Domain

We can obtain our general current equation using nodal analysis from our parallel RLC circuit
above,

 
1 1
V i(s) = V o(s) sC + +
sL R
 2 
s RLC + sL + R
V i(s) = V o(s) (1)
sRL
Considering the current at capacitor we get,
IC (s) = V o(s) × sC
IC (s)
V o(s) =
sC
We substitute the value of V o(s) to equation 1,
IC (s) s2 RLC + sL + R
 
V i(s) =
sC sRL
 2 
s RLC + sL + R
V i(s) = IC (s)
s2 RLC
Thus, the transfer function we get with the current from capacitor as our output is,
IC (s) s2 RLC s2
= 2 = 2 1 1 (2)
V i(s) s RLC + sL + R s + s RC + LC

11
Considering the current at inductor we get,

V o(s)
IL (s) =
sL
V o(s) = IL (s) × sL

Substituting the value of V o(s) to equation 1,

s2 RLC + sL + R
 
V i(s) = IL (s) × sL
sRL
 2 
s RLC + sL + R
V i(s) = IL (s)
R

Thus, our transfer function with the inductor current as our output is,
1
IL (s) R LC
= 2 = 2 1 1 (3)
V i(s) s RLC + sL + R s + s RC + LC

Considering the current at resistor we get,

V o(s)
IR (s) =
R
V o(s) = IR (s) × R

Substituting the value of V o(s) to equation 1,

 2 
s RLC + sL + R
V i(s) = IR (s) × R
sRL
 2 
s RLC + sL + R
V i(s) = IR (s)
sL

Thus, our transfer function with the resistor current as our output is,
1
IR (s) sL s RC
= 2 = 2 1 1 (4)
V i(s) s RLC + sL + R s + s RC + LC

6.2 Series RLC Circuit

Figure 8: Series RLC in Frequency Domain

12
We can obtain the general voltage equation using mesh analysis,
 
1
V i(s) = I(s) R + sL +
sC
 2 
s LC + sRC + 1
V i(s) = I(s) (5)
sC
For the output voltage from the resistor,
VR (s) = I(s) × R
VR (s)
I(s) =
R
Substitute it to equation 5,
s2 LC + sRC + 1
 
VR (s)
V i(s) =
R sC
 2 
s LC + sRC + 1
V i(s) = VR (s)
sRC
Thus, we can get the transfer function,
VR (s) sRC sR
L
= 2 = 2 (6)
V i(s) s LC + sRC + 1 s + sR
L
+ 1
LC

For the output voltage from the inductor,

VL (s) = I(s) × sL
VL (s)
I(s) =
sL
Substituting to equation 5,
s2 LC + sRC + 1
 
VL (s)
V i(s) =
sL sC
 2 
s LC + sRC + 1
V i(s) = VL (s)
s2 LC
We can obtain the transfer function,
VL (s) s2 LC s2
= 2 = 2 (7)
V i(s) s LC + sRC + 1 s + sR
L
+ 1
LC

For the output voltage at the capacitor,

I(s)
VC (s) =
sC
I(s) = VC (s) × sC
Substituting to equation 5,
s2 LC + sRC + 1
 
V i(s) = VC (s) × sC
sC
 2 
s LC + sRC + 1
V i(s) = VL (s)
1
Thus, we can get the transfer function,
1
VC (s) 1 LC
= 2 = 2 (8)
V i(s) s LC + sRC + 1 s + sR
L
+ 1
LC

13
7 Time Domain
To obtain the time domain functions of the previous transfer functions, we have utilized
ilaplace() command of MATLAB.

7.1 Parallel RLC Circuit

Getting the time domain transfer function for the output at:
i. Capacitor
syms s t R L C;
Ic = (s^2)/(s^2+ s*(1/(R*C)) + (1/(L*C)))
ic = ilaplace(Ic)

and the resulting time domain function,

 √ ! 
√ −CR2 + L
√  C LR sinh
t
√ 4
( R 1
− 2CR )
L
t −CR2 + L C LR
t √


exp − 2CR cosh √ 4
+

C LR −CR2 + L

4

iC (t) = δ(t) −
CR
(9)

ii. Inductor
syms s t R L C;
Il = (1/(L*C))/(s^2+ s*(1/(R*C)) + (1/(L*C)))
il = ilaplace(Il)

and obtain the transfer funnction of,

√ 
 t

t −4CR 2 +L
2R exp − 2CR sinh √
2C LR
iL (t) = √ √ (10)
2
L −4CR + L

iii. Resistor
syms s t R L C;
Ir = (s/(R*C))/(s^2+ s*(1/(R*C)) + (1/(L*C)))
ir = ilaplace(Ir)

obtains the transfer function,

 √ !
√ −CR2 + L
√  L sinh
t
√ 4
t −CR2 + L C LR
t √


exp − 2CR cosh √ 4
−

C LR −CR2 + L
2

4

iR (t) = (11)
CR

7.2 Series RLC Circuit

Obtaining the time domain transfer function for the voltage output at:
i. Resistor

14
 syms s t R L C;
Vr = (s*R*C)/((s^(2)*L*C)+(s*R*C)+1)
vr = ilaplace(Vr)

yields to the transfer function

 √ !
√ t CR2 −L
CR sinh √4
 q 
2
Rt 

t CR −L CL
R exp − 2L cosh √ 4
− q

CL 2 
2 CR
4
−L

vr (t) = (12)
L

ii. Inductor
syms s t R L C;
Vl = (s^(2)*L*C)/((s^(2)*L*C)+(s*R*C)+1)
vl = ilaplace(Vl)

obtains the transfer function,

 √ ! 
√ t CR2 −L
R 1
 q
t CR
2
−L
 CL sinh √4
CL
( 2L − CR )
Rt 


R exp − 2L cosh √ 4
− q

CL CR2

4
−L

vL (t) = δ(t) −
L
(13)
iii. Capacitor
syms s t R L C;
Vc = (1)/((s^(2)*L*C)+(s*R*C)+1)
vc = ilaplace(Vc)

obtains the transfer function,

 √ 
t −CR 2
√ +4L exp Rt


2 sin 2 CL
− 2L
vC (t) = √ √ (14)
C −CR2 + 4L

8 State Space Model

We define the transfer function that relates to the input and output from the state space
[Link] also represent the state space model into its phase variable, controller canonical and
observer canonical forms of the state space model.
State space representation is defined by

ẋ(t) = Ax(t) + Bu(t) (15)

y(t) = Cx(t) + Du(t) (16)

15
8.1 Parallel RLC Circuit

For parallel RLC circuit, we can identify our state variables as the voltage across the capacitor
and the current across the inductor.
   
vC x1 (t)
x= =
iL x2 (t)
We obtain our first state space equation from the current at capacitor,
dvC (t)
iC (t) = C
dt
where

iC (t) = iT (t) − iL (t) − iR (t)

dvC (t)
C = iT (t) − iL (t) − iR (t)
dt
dvC (t) vC (t)
C = vi (t) − vC (t) − iL (t) −
dt R
dvC (t) 1 R+1 1
= vi (t) − vC (t) − iL (t)
dt C RC C
1 R+1 1
ẋ1 (t) = vi (t) − x1 (t) − x2 (t) (17)
C RC C
Then we obtain our second state space equation from the voltage across the inductor,
diL (t)
L = vC (t)
dt
diL (t) 1
= vC (t)
dt L
1
ẋ2 (t) = x1 (t) (18)
L
And our output equations for each of the elements are:
i. Capacitor

yC (t) = iT (t) − iL (t) − iR (t)

1
yC (t) = vi (t) − vC (t) − iL (t) − vC (t)
R
R+1
yC (t) = vi (t) − vC (t) − iL (t)
R
R+1
yC (t) = vi (t) − x1 (t) − x2 (t) (19)
R
16
 ii. Inductor

yL (t) = iL (t)
yL (t) = x2 (t) (20)

iii. Resistor
1
yR (t) = vC (t)
R
1
yR (t) = x1 (t) (21)
R
Thus, we obtain the state space system representation of
  R+1  1
− RC − C1
  
ẋ1 (t) x1 (t)
= 1 · + C · u(t)
ẋ2 (t) L
0 x 2 (t) 0
 
 R+1  x1 (t)  
yC (t) = − R −1 · + 1 · u(t)
x2 (t)
 
  x1 (t)  
yL (t) = 0 1 · + 0 · u(t)
x2 (t)
 
1  x1 (t)  
yR (t) = R 0 · + 0 · u(t)
x2 (t)

8.1.1 Phase Variable Form

We consider the transfer function follows

Y (s) b1 s + b2
G(s) = = 2 (22)
G(s) s + a1 s + a2

and state variable as

x1 (t) = u̇(t); x2 (t) = u(t) (23)

to get
ẋ(t) = A2 x(t) + B2 u(t) (24)
where    
−a1 −a2 1
A2 = ; B2 =
0 1 0
Then, consider that

y(t) = b1 ẋ(t) + b2 x(t)

(25)
y(t) = Cx(t) + Du(t)

where    
C = b 1 b2 ; D = 0

17
i. at Capacitor

The state space representation in time domain is

   1 1
    
ẋ1 (t) − RC − LC x1 (t) 1
= · + · u(t)
ẋ2 (t) 0 1 x2 (t) 0
 
  x1 (t)  
yC (t) = 0 0 · + 1 · u(t)
x2 (t)
We take the Laplace transform of the previous equations and get the state space
equation in phase variable form as
   1 1
    
Ẋ1 (s) − RC − LC X1 (s) 1
= · + · U (s)
Ẋ2 (s) 0 1 X2 (s) 0

 X1 (s)
  
YC (s) = 0 0 · + 1 · U (s)
X2 (s)
The state space in phase variable form represented in a signal flow graph

18
ii. Inductor

The state space representation in time domain is

   1 1
    
ẋ1 (t) − RC − LC x1 (t) 1
= · + · u(t)
ẋ2 (t) 0 1 x2 (t) 0
 
 1
 x1 (t)  
yL (t) = 0 LC · + 0 · u(t)
x2 (t)
We take the Laplace transform of the previous equations and get the state space
equation in phase variable form as
   1 1
    
Ẋ1 (s) − RC − LC X1 (s) 1
= · + · U (s)
Ẋ2 (s) 0 1 X2 (s) 0
 
 1
 X1 (s)  
YL (s) = 0 LC
· + 0 · U (s)
X2 (s)
The state space in phase variable form represented in a signal flow graph

19
 iii. Resistor

The state space representation in time domain is

   1 1
    
ẋ1 (t) − RC − LC x1 (t) 1
= · + · u(t)
ẋ2 (t) 0 1 x2 (t) 0
 
 1
 x1 (t)  
yR (t) = 0 RC · + 0 · u(t)
x2 (t)
We take the Laplace transform of the previous equations and get the state space
equation in phase variable form as
   1 1
    
Ẋ1 (s) − RC − LC X1 (s) 1
= · + · U (s)
Ẋ2 (s) 0 1 X2 (s) 0
 
 1
 X1 (s)  
YR (s) = RC
0 · + 0 · U (s)
X2 (s)
The state space in phase variable form represented in a signal flow graph

8.1.2 Controller Canonical Form

i. Capacitor        
Ẋ1 (s) 0 1 X1 (s) 0
= 1 1 · + · U (s)
Ẋ2 (s) − LC − RC X2 (s) 1
 
  X1 (s)  
Y (s) = 0 0 · + 1 · U (s)
X2 (s)
The controller canonical form is represented by the signal flow graph

20
ii. Inductor        
Ẋ1 (s) 0 1 X1 (s) 0
= 1 1 · + · U (s)
Ẋ2 (s) − LC − RC X2 (s) 1
 
 1  X1 (s)  
Y (s) = LC 0 · + 0 · U (s)
X2 (s)
The controller canonical form is represented by the signal flow graph

iii. Resistor        
Ẋ1 (s) 0 1 X1 (s) 0
= 1 1 · + · U (s)
Ẋ2 (s) − LC − RC X2 (s) 1
 
 1
 X1 (s)  
Y (s) = 0 RC · + 0 · U (s)
X2 (s)
The controller canonical form is represented by the signal flow graph

21
8.1.3 Observer Canonical Form

The relationship between state space matrices in controller and observer canonical form are

Aobs = ATcont T
Bobs = Bcont
T T
Cobs = Ccont Dobs = Dcont

i. Capacitor
We have state space equation in observer canonical form as
1
       
Ẋ1 (s) 0 − LC X1 (s) 0
= 1 · + · U (s)
Ẋ2 (s) 1 − RC X2 (s) 0
 
  X1 (s)  
Y (s) = 0 1 · + 1 · U (s)
X2 (s)
The observer canonical form is represented by the signal flow graph

ii. Inductor
We have state space equation in observer canonical form as
1
       
Ẋ1 (s) 0 − LC X1 (s) 0
= 1 · + 1 · U (s)
Ẋ2 (s) 1 − RC X2 (s) LC
 
  X1 (s)  
Y (s) = 0 1 · + 0 · U (s)
X2 (s)
The observer canonical form is represented by the signal flow graph

22
 iii. Resistor
We have state space equation in observer canonical form as
1
       1 
Ẋ1 (s) 0 − LC X1 (s)
= 1 · + RC · U (s)
Ẋ2 (s) 1 − RC X2 (s) 0
 
  X1 (s)  
Y (s) = 0 1 · + 0 · U (s)
X2 (s)
The observer canonical form is represented by the signal flow graph

8.2 Series RLC Circuit

For series RLC, the state space variables are the same with the state variables from a parallel
RLC circuit. Which is the current across the inductor and the voltage across the capacitor.
   
i x1 (t)
x= =
vc x2 (t)

23
We can obtain the first state space equation by using mesh analysis,

di(t)
L = vi(t) − Ri(t) − vC (t)
dt
di(t) 1 R 1
= vi(t) − i(t) − vC (t)
dt L L L
1 R 1
ẋ1 (t) = vi(t) − x1 (t) − x2 (t) (26)
L L L
Another state space equation is by equating the current at capacitor,

dvC (t)
i(t) = C
dt
dvC (t) 1
= i(t)
dt C
1
ẋ2 (t) = x1 (t) (27)
C
Then, we obtain the output equations for each of the elements,
i. Resistor

yR (t) = Ri(t)
yR (t) = Rx1 (t) (28)

ii. Inductor

yL (t) = Li(t)
yL (t) = Lx1 (t) (29)

iii. Capacitor

yC (t) = vC (t)
yC (t) = x2 (t) (30)

Thus, we obtain the state space system representation of

  R  1
− L − L1
  
ẋ1 (t) x1 (t)
= 1 · + L · u(t)
ẋ2 (t) C
0 x 2 (t) 0
 
  x1 (t)  
yR (t) = R 0 · + 0 · u(t)
x2 (t)
 
  x1 (t)  
yL (t) = L 0 · + 0 · u(t)
x2 (t)
 
  x1 (t)  
yC (t) = 0 1 · + 0 · u(t)
x2 (t)

24
8.2.1 Phase Variable Form
i. at Resistor

The state space representation in time domain is

   R 1
    
ẋ1 (t) − L − LC x1 (t) 1
= · + · u(t)
ẋ2 (t) 0 1 x2 (t) 0
 
 R  x1 (t)  
yR (t) = L 0 · + 0 · u(t)
x2 (t)
We take the Laplace transform of the previous equations and get the state space
equation in phase variable form as
   R 1
    
Ẋ1 (s) − L − LC X1 (s) 1
= · + · U (s)
Ẋ2 (s) 0 1 X2 (s) 0
 
R  X1 (s)  
YR (s) = L
0 · + 0 · U (s)
X2 (s)
The state space in phase variable form represented in a signal flow graph

ii. Inductor

The state space representation in time domain is

   R 1
    
ẋ1 (t) − L − LC x1 (t) 1
= · + · u(t)
ẋ2 (t) 0 1 x2 (t) 0

25
  
  x1 (t)  
yL (t) = 0 0 · + 1 · u(t)
x2 (t)
We take the Laplace transform of the previous equations and get the state space
equation in phase variable form as
   R 1
    
Ẋ1 (s) − L − LC X1 (s) 1
= · + · U (s)
Ẋ2 (s) 0 1 X2 (s) 0
 
  X1 (s)  
YL (s) = 0 0 · + 1 · U (s)
X2 (s)
The state space in phase variable form represented in a signal flow graph

iii. Capacitor

The state space representation in time domain is

   R 1
    
ẋ1 (t) − L − LC x1 (t) 1
= · + · u(t)
ẋ2 (t) 0 1 x2 (t) 0
 
 1
 x1 (t)  
yC (t) = 0 LC · + 0 · u(t)
x2 (t)
We take the Laplace transform of the previous equations and get the state space
equation in phase variable form as
   R 1
    
Ẋ1 (s) − L − LC X1 (s) 1
= · + · U (s)
Ẋ2 (s) 0 1 X2 (s) 0
 
 1
 X1 (s)  
YC (s) = 0 LC
· + 0 · U (s)
X2 (s)
The state space in phase variable form represented in a signal flow graph

26
8.2.2 Controller Canonical Form
i. Resistor        
Ẋ1 (s) 0 1 X1 (s) 0
= 1 R · + · U (s)
Ẋ2 (s) − LC − L X2 (s) 1
 
 R  X1 (s)
Y (s) = 0 L · +] ·[ U (s)
X2 (s)
The controller canonical form is represented by the signal flow graph

ii. Inductor        
Ẋ1 (s) 0 1 X1 (s) 0
= 1 R · + · U (s)
Ẋ2 (s) − LC − L X2 (s) 1
 
  X1 (s)  
Y (s) = 0 0 · + 1 · U (s)
X2 (s)
The controller canonical form is represented by the signal flow graph

27
 iii. Capacitor        
Ẋ1 (s) 0 1 X1 (s) 0
= 1 R · + · U (s)
Ẋ2 (s) − LC − L X2 (s) 1
 
 1  X1 (s)  
Y (s) = LC 0 · + 0 · U (s)
X2 (s)
The controller canonical form is represented by the signal flow graph

8.2.3 Observer Canonical Form

The relationship between state space matrices in controller and observer canonical form are

Aobs = ATcont T
Bobs = Bcont
T T
Cobs = Ccont Dobs = Dcont

i. Resistor
We have state space equation in observer canonical form as
1
      R
Ẋ1 (s) 0 − LC X1 (s)
= · + L · U (s)
Ẋ2 (s) 1 −R L
X 2 (s) 0
 
  X1 (s)  
Y (s) = 0 1 · + 0 · U (s)
X2 (s)
The observer canonical form is represented by the signal flow graph

28
ii. Inductor
We have state space equation in observer canonical form as
1
       
Ẋ1 (s) 0 − LC X1 (s) 0
= R · + · U (s)
Ẋ2 (s) 1 − L
X 2 (s) 0
 
  X1 (s)  
Y (s) = 0 1 · + 1 · U (s)
X2 (s)
The observer canonical form is represented by the signal flow graph

iii. Capacitor
We have state space equation in observer canonical form as
1
       
Ẋ1 (s) 0 − LC X1 (s) 0
= R · + 1 · U (s)
Ẋ2 (s) 1 −L X2 (s) LC
 
  X1 (s)  
Y (s) = 0 1 · + 0 · U (s)
X2 (s)
The observer canonical form is represented by the signal flow graph

29