CL3601 – Process Control

Assignment 1 – Process Dynamics

QUESTION 1

A.

STUDENT ID: 2154778

𝑥 = 8+1=9

Now using the given transfer function:

𝑌(𝑠) 12(𝑠 + 4)
𝐺(𝑠) = =
𝑈(𝑠) (𝑠 + 9)(𝑠 + 3)(𝑠 + 8)

For a system represented by a transfer function, if all poles (the values of s that make
the denominator zero) have negative real parts, the system will reach a steady state
in response to a step input.

The poles are the values of ss that make the denominator zero. So, we can find the poles by
setting:
(s+9) (s+3) (s+8) = 0

From this equation, the poles are s = −9, −3 and −8.

All of these poles have negative real parts. Hence, the system will reach a steady state in
response to a step input.

Conclusion: Yes, if a step change is made to the reactor temperature, the mass fraction of
A in the product will reach a steady value. This is because all the poles of the transfer
function have negative real parts, which ensures system stability in response to a step input.

B.
Given that we already have:
𝑌(𝑠)
𝐺(𝑠) =
𝑈(𝑠)

We need to relate U(s) (reactor temperature) to Z(s) (coolant flow rate) through G2(s).
The given relationship is:
𝑈(𝑠) = 𝐺! (𝑠) × 𝑍(𝑠)

To derive the relationship between Y(s) and Z(s), you multiply G1(s) with U(s):

𝑌(𝑠) = 𝐺" (𝑠) × 𝑈(𝑠)

Substituting 𝑈(𝑠) = 𝐺! (𝑠) × 𝑍(𝑠) into the above equation:

𝑌(𝑠) = 𝐺" (𝑠) × 𝐺! (𝑠) × 𝑍(𝑠)

Thus, the relationship (or transfer function) between Y(s) and Z(s) is:

𝐺(𝑠) = 𝐺" (𝑠) × 𝐺! (𝑠)

This transfer function G(s) describes how the mass fraction of product A leaving the reactor
Y(s) relates to the coolant flow rate Z(s).

Note: The actual form of G2(s) was not provided in your question, so I couldn't compute the
exact G(s). But the general methodology and formula for combining them is as described
above.

C.

Given
12(𝑠 + 4)
𝐺(𝑠) =
(𝑠 + 9)(𝑠 + 3)(𝑠 + 8)

Chose K=1 for simplicity. The exact value of K doesn't affect system stability unless it leads
to a non-minimum phase zero, but this is not the case here.

Using the provided structure for G2(s):

𝐾
𝐺! (𝑠) =
𝜏𝑠 + 1
The combined transfer function becomes:
12(𝑠 + 4)𝐾
𝐺(𝑠) =
(𝑠 + 9)(𝑠 + 3)(𝑠 + 8)(𝜏𝑠 + 1)

Stable System Example:

Choose K=1 and τ=5 for simplicity:

12(𝑠 + 4) × 1
𝐺(𝑠) =
(𝑠 + 9)(𝑠 + 3)(𝑠 + 8)(5𝑠 + 1)
All poles are in the left-half plane (negative), ensuring the system is stable. Thus, a step
change in the coolant flow rate will lead the mass fraction of product A to reach a steady
value.

Unstable System Example:

To create an unstable system, we can pick τ to be negative, say τ = −1:

12(𝑠 + 4) × 1
𝐺(𝑠) =
(𝑠 + 9)(𝑠 + 3)(𝑠 + 8)(−𝑠 + 1)

Here, the pole s=1 ensures the system is unstable. For a unit step change in the coolant flow
rate, the value of A will not settle to a steady value but will grow without bound.
QUESTION 2

Figure 1: diagram of the system

Mass Balance:

For a steady-state condition, the mass entering the tank equals the mass leaving the tank.
. .
𝑚#$ = 𝑚&'(

Given the constant height of the liquid, let's compute the mass of dodecane in the tank.
The volume V of the dodecane in the tank is given by:

𝑑 !
𝑉 =𝜋×9 ; ×ℎ
2

Substituting for d=2 m and h=6.5 m:

𝑉 = 𝜋 × (1)! × 6.5

𝑉 ≈ 20.42 𝑚)

Using the given density ρ=750 kg/m3, the mass m of dodecane in the tank is:

𝑚 =𝜌×𝑉

𝑚 = 750 × 20.42

𝑚 ≈ 15,315𝑘𝑔
Energy Balance:

The energy required to raise the temperature of the dodecane from the inlet temperature to
the tank temperature is given by:
𝑄 = 𝑚 × 𝐶* × ∆𝑇
Where:
∆𝑇 = 𝑇(+$, − 𝑇#$-.(

∆𝑇 = 150℃ − 10℃ = 140℃

Using the given specific heat 𝐶* = 2200𝐽/𝑘𝑔/ 𝐾 /

𝑄 = 15,315 × 2200 × 140

𝑄 ≈ 4.709 × 100 𝐽
Summary:

• The mass of dodecane in the tank at steady-state is approximately 15,315 kg.

• The amount of heat required to heat the dodecane from 10°C to 150°C in the tank is
approximately 4.709 × 100 𝐽 Joules.
•
(Note: Assumptions made include steady-state conditions, constant density and specific heat
of dodecane, and perfect insulation of the tank.)

B.

Given:

ℎ
𝐹&'( =
𝑅

When the system is at steady state, the inlet flow rate Fin equals the outlet flow rate Fout.
This means:

𝐹#$ = 𝐹&'(

Given 𝐹#$ =120 L/min, and from the previous question, we have the liquid height h=6.5 m.
Substituting the given values into the equation:
6.5𝑚
120𝐿/𝑚𝑖𝑛 =
𝑅
To solve for R:
6.5𝑚
𝑅=
120𝐿/𝑚𝑖𝑛

𝑅 ≈ 0.0542𝑚/𝐿/𝑚𝑖𝑛

Therefore, the resistance of the outlet valve R is approximately 0.0542 m/L/min.

C.

Given:

• Inlet temperature, 𝑇#$-.( = 10°C

• Outlet temperature, 𝑇&'( = 150°C
• Flow rate, 𝐹#$-.( = 120 L/min = 0.12 𝑚) /min
• Specific heat capacity of dodecane, Cp = 2200 J/kg·K
• Density of dodecane, ρ = 750 kg/m^3

The mass flow rate, m˙ is given by:

𝑚˙ = 𝜌 × 𝐹#$-.(

𝑚˙ = 750𝑘𝑔/𝑚) × 0.12𝑚) /𝑚𝑖𝑛

𝑚 = 90𝑘𝑔/𝑚𝑖𝑛

The energy required per minute to heat the dodecane, Q˙, is given by:

Q˙= 𝑚˙ × 𝐶* × ∆𝑇

Where ΔT is the change in temperature = 𝑇&'(-.( − 𝑇#$-.(

ΔT = 150°C − 10°C = 140°C

Substitute in the values:
𝑄˙ = 90𝑘𝑔/𝑚𝑖𝑛 × 2200𝐽/𝑘𝑔. 𝐾 × 140𝐾

𝑄˙ = 90 × 2200 × 140

𝑄˙ = 277,200,000𝐽/𝑚𝑖𝑛

𝑄˙ = 277,2𝑀𝐽/𝑚𝑖𝑛

So, the energy required to be supplied by the heater such that the outlet temperature is
150°C is 277.2 MJ/min.

D.

Steady-State Value of the Liquid Height:

The steady-state height is when the input flow rate equals the output flow rate. From part b,
we have:
ℎ
𝐹&'( =
𝑅
Using the new inlet flow rate:
𝐹#$ = 180𝐿/𝑚𝑖𝑛 = 0.18𝑚) /𝑚𝑖𝑛

For steady-state:
𝐹#$ = 𝐹&'(

ℎ
0.18 =
5.4
 ℎ = 0.972𝑚
The height increases by 0.972 m.

Steady-State Value of the Liquid Temperature:

The energy balance equation in steady-state is:

Q˙= 𝑚˙ × 𝐶* × ∆𝑇

Q˙= 550 kW =550,000 J/min

𝑚˙ = 𝜌 × 𝐹#$-.(

𝑚˙ = 750𝑘𝑔/𝑚) × 0.18𝑚) /𝑚𝑖𝑛

𝑚 = 135𝑘𝑔/𝑚𝑖𝑛

The energy required per minute to heat the dodecane, Q˙, is given by:

Q˙= 𝑚˙ × 𝐶* × ∆𝑇

Where ΔT is the change in temperature = 𝑇&'(-.( − 𝑇#$-.(

ΔT = 𝑇&'(-.( − 10°C
Substitute in the values:

550,000 = 135𝑘𝑔/𝑚𝑖𝑛 × 2200𝐽/𝑘𝑔. 𝐾 × ( 𝑇𝑜𝑢𝑡𝑙𝑒𝑡 − 10°𝐶)

550,000 = 135 × 2200 × ( 𝑇𝑜𝑢𝑡𝑙𝑒𝑡 − 10°𝐶)

𝑇&'(-.( = 176.85℃

Gain and Time Constant:

The gain (K) is defined as the change in output over the change in input. Given a stepwise
change in height of 0.972 m (from our previous calculations) due to an input change of 60
L/min:

∆𝑜𝑢𝑡𝑝𝑢𝑡
𝐾=
∆𝑖𝑛𝑝𝑢𝑡

0.972
𝐾= = 16.2𝑚) /𝐿
60 × 10/)

The time constant (τ) can be found using the tank volume and the flow rate:

𝑣𝑜𝑙𝑢𝑚𝑒 𝜋𝐷! ℎ/4

𝜏= =
𝑓𝑙𝑜𝑤𝑟𝑎𝑡𝑒 𝐹#$

Given h=7.47 m and D=2 m:

𝜋 × 2! × 7.47/4
𝜏= = 57.42 𝑚𝑖𝑛
0.18
PLOT:

QUESTION 4

1 . Basic differential relationship of liquid volume in a tank to flow rates.

𝑑𝑉
= 𝐹#$," − 𝐹"
𝑑𝑇

2 . Relate volume V to the height h1 of the liquid in the tank.

𝑉 = 𝐴 × ℎ"

Where A is the cross-sectional area of the tank.

Differentiating with respect to time,
𝑑𝑉 𝑑ℎ"
=𝐴×
𝑑𝑡 𝑑𝑡

3 . From the given information, the outlet flow rate F1 is given as:
ℎ"
𝐹" =
𝑅"

Substitute the equations from Step 2 and Step 3 into the differential equation from
Step 1:
𝑑ℎ" ℎ"
𝐴× = 𝐹#$," −
𝑑𝑡 𝑅"

Take the Laplace Transform of the above equation (assuming all initial conditions are zero
for simplicity):
𝐻" (𝑠)
𝐴 × 𝑠𝐻" (𝑠) = 𝐹#$," (𝑠) −
𝑅"
Where H1(s) is the Laplace transform of h1(t) and Fin,1(s) is the Laplace transform of
Fin,1(t).

7! (9)
Rearrange the equation to solve for the transfer function 𝐺(𝑠) × ; :
"#,! (9)
𝐻" (𝑠) 𝐴×𝑠
𝐺(𝑠) × =
𝐹#$," (𝑠) 𝑠𝐴 + 1/𝑅"

This is the transfer function that relates the height h1 in Tank #1 to its inlet flow Fin,1.

B.

<!
For this question, we will use the equation 𝐹" =
=!

Given:

1. h1=1 m
2. R1=200 s/m^2

Substitute in the values:

">
𝐹#$," = !??9/>%

𝐹#$," = 0.005 𝑚/𝑠

Thus, the flow rate of liquid entering Tank #1 is Fin,1=0.005 m/s.

C.

To calculate the gain and time constants for Tank #1, let's derive the transfer function. The
transfer function G(s) relates the Laplace Transform of the output (height h1) to the Laplace
Transform of the input (inlet flow Fin,1).
From the given model equations:
ℎ" 𝑑ℎ"
𝐹#$," (𝑠) − = 𝐴"
𝑅" 𝑑𝑡
Taking the Laplace Transform:
𝐻" (𝑠)
𝐹#$," (𝑠) − = 𝐴" 𝑠𝐻" (𝑠)
𝑅"
7! (9)
Rearranging and solving for ; :
"#,! (9)
𝐻" (𝑠) 𝑅"
=
𝐹#$," (𝑠) 𝐴" 𝑠 + 𝑅"
The transfer function G(s) is given by:
𝐻" (𝑠) 𝑅"
𝐺(𝑠) = =
𝐹#$," (𝑠) 𝐴" 𝑠 + 𝑅"

For a first-order system, the transfer function can be represented as:

𝐾
𝐺(𝑠) =
𝜏𝑠 + 1
Where:
• K is the steady-state gain.
• τ is the time constant.
Comparing the two forms of the transfer function:
𝐾 = 𝑅"
𝜏𝐴" = 𝑅"

From our previous discussions:

𝑅" = 200𝑠/𝑚!
𝐴" = 0.19635𝑚!
Thus:
K = 200
𝑅" 200𝑠/𝑚!
𝜏= = ≈ 1019.4𝑠
𝐴" 0.19635𝑚!
In conclusion:
• The gain K for Tank #1 is 200.
• The time constant τ for Tank #1 is approximately 1019.4 seconds.

D.
1. Formulate the Equations for Each Tank:
For Tank #1:
The rate of change of height h1 is proportional to the difference between the inflow and
outflow.
𝑑ℎ"
𝐴" = 𝐹#$," − 𝐹"
𝑑𝑡
𝑑ℎ" ℎ"
𝐴" = 𝐹#$," −
𝑑𝑡 𝑅"
For Tank #2:
Similarly, the rate of change of height h2 is proportional to the difference between the inflow
and outflow.
𝑑ℎ!
𝐴! = 𝐹#$,! − 𝐹!
𝑑𝑡
𝑑ℎ! ℎ!
𝐴! = 𝐹#$,! −
𝑑𝑡 𝑅!
For Tank #3:
The rate of change of height h3 is proportional to the sum of outflows from Tanks #1 and #2
minus the outflow from Tank #3.

𝑑ℎ)
𝐴) = 𝐹" + 𝐹! − 𝐹)
𝑑𝑡
𝑑ℎ) ℎ" ℎ! ℎ)
𝐴) = + −
𝑑𝑡 𝑅" 𝑅! 𝑅)

2. Take the Laplace Transform:

Using the property that the Laplace transform of the derivative is s× transform of the
function minus its initial value, and assuming zero initial conditions:
For Tank #1
𝐻" (𝑠)
𝐴" 𝑠𝐻" (𝑠) = 𝐹#$," (𝑠) −
𝑅"
For Tank #2:
𝐻! (𝑠)
𝐴! 𝑠𝐻! (𝑠) = 𝐹#$,! (𝑠) −
𝑅!
For Tank #3:
𝐻) (𝑠)
𝐴) 𝑠𝐻) (𝑠) = 𝐹#$,) (𝑠) −
𝑅)

For Tank #3, based on the given equations, the dynamics in the s-domain are given by:

𝐻" (𝑠) 𝐻! (𝑠) 𝐻) (𝑠)

𝐴) 𝑠𝐻) (𝑠) = + −
𝑅" 𝑅! 𝑅)

Rearranging the equation for Tank #3 to isolate H3(s):

𝐻) (𝑠) 𝐻" (𝑠) 𝐻! (𝑠)

𝐴) 𝑠𝐻) (𝑠) + = +
𝑅) 𝑅" 𝑅!

1 𝐻" (𝑠) 𝐻! (𝑠)

𝐻) (𝑠)(𝐴) 𝑠 + )= +
𝑅) 𝑅" 𝑅!

Using the expressions for H1(s) and H2(s) derived earlier:

1 𝐴" 𝑠𝐹#$," (𝑠) − 𝐹#$," (𝑠) 𝐴! 𝑠𝐹#$,! (𝑠) − 𝐹#$,! (𝑠)

𝐻) (𝑠)(𝐴) 𝑠 + )= +
𝑅) 𝑅" 𝑅!

This equation gives the relationship between H3(s), Fin,1(s), and Fin,2(s). However, to
express this as a transfer function relating an input to the output H3(s), we would need to
rearrange the equation with respect to either Fin,1(s) or Fin,2(s).

For example, if we want the transfer function from Fin,1(s) to H3(s).

𝐴" 𝑠 − 1
𝐻) (𝑠) 𝑅"
=
𝐹#$," (𝑠) 𝐴 𝑠 + 1 − 𝐴! 𝑠 − 1
) 𝑅) 𝑅!

Similarly, for Fin,2(s) to H3(s):

𝐴! 𝑠 − 1
𝐻) (𝑠) 𝑅!
=
𝐹#$,! (𝑠) 𝐴 𝑠 + 1 𝐴" 𝑠 − 1
) 𝑅 − 𝑅) "
Both of these transfer functions give the relationship between the respective input flow rates
and the height of the liquid in Tank #3.
QUESTION 3

1. How the changes to the design affect the mass and energy balances for the
system.

By removing the insulation:

• The system will have an increased rate of heat loss to the surroundings
because there's no insulation to reduce the heat transfer.
• Due to the increased heat loss, more energy would be required to maintain
the tank at the desired temperature of 150 °C.
• The mass balance won't change, but the energy balance will be affected
because of the change in the heat loss rate.

2. Determine the energy required for the uninsulated tank:

Heat Loss due to no insulation:

𝑄-&99 = ℎ × 𝐴 × (𝑇(+$, − 𝑇+#A )

Using the given:

ℎ = 20𝑊𝑚/! 𝐾 /

𝑇(+$, = 150℃ = 423𝐾

𝑇+#A = 10℃ = 283𝐾

(For the area, A, we would ideally consider both the cylindrical side and the two ends of the
tank, but since the area is not provided, we cannot compute this.)

So, 𝑄-&99 = 20 × 𝐴 × (423 − 283)

𝑄-&99 = 20 × 𝐴 × (140)

𝑄-&99 = 2800𝐴𝑊

This is the energy loss per unit time due to no insulation.

3.For the stepwise increase in the inlet flow rate from 120 to 180 L min⁻¹:

Let's consider an arbitrary specific heat for dodecane (not provided) as 𝐶B The energy
required to heat the increased flow rate is:

∆Q = ∆𝑚 × 𝐶* × ∆𝑇
Where:

∆𝑚 = increased mass flow rate (this would be the difference in flow rate x density of
dodecane, which again isn't provided).

∆𝑇 = temperature difference to be maintained, which is 150°C - 10°C = 140°C.

For the uninsulated tank, the additional energy from the heater would be:

𝑄(&(+- = 𝑄-&99 + ∆𝑄
3. Comparison:

For the insulated system, the heat loss 𝑄-&99 would be significantly lesser than the
uninsulated one. Thus, for the same stepwise increase in inlet flow rate:

𝑄(&(+-,#$9'-+(.C = ∆𝑄

Comparing the two:

𝑄(&(+-,'$#$9'-+(.C = 2800 × 𝐴 + ∆𝑚 × 𝐶B × 140

𝑄(&(+-,#$9'-+(.C = ∆𝑚 × 𝐶B × 140

From the equations, it's evident that the uninsulated system demands more energy due to
the added 2800×A term, which is the heat loss to the surroundings.

Conclusion: Building the system without insulation would require a significantly higher
energy input, especially for higher flow rates. Therefore, not insulating the system would be
less efficient and could be more expensive in the long run.
Note: The above analysis is a theoretical approach. Actual values would require more
specific data like the tank's surface area, dodecane's density, and specific heat.