CHE3007S Block3

Lecture 4
Developing transfer functions
Overview
• What are transfer functions?
• What are the properties of a transfer function?
• What is a first order response?
Additional Reading / Learning
Sections of textbook by Seborg et al. to study:
• Chapter 3 – 3rd Editions
• Chapter 4 – 4th Editions

Glossary:
• Transfer function, Deviation variable, Transfer function order, Time
constant, Process gain
What are transfer functions?
• A transfer function is an algebraic expression for the dynamic relation
between a selected input and output of a process.

O(s) – Output
𝑂(𝑠) = 𝑮 𝒔 𝐼(𝑠) G(s) – Transfer function
I(s) – Input

• A transfer function is defined in such a way that it is independent of

the initial conditions, the function of the disturbance and or the
manipulated variable i.e. it is general.
What are transfer functions?
• A transfer function is an algebraic expression for the dynamic relation
between a selected input and output of a process.
Convolution
𝐼(𝑡) 𝐺(𝑡) 𝑂 𝑡 = 𝐼 𝑡 ∗ 𝐺(𝑡) integral

𝐼(𝑠) 𝑮(𝒔) 𝑂 𝑠 = 𝐼 𝑠 𝑮(𝒔)

• A transfer function is defined in such a way that it is independent of

the initial conditions, the function of the disturbance and or the
manipulated variable i.e. it is general.
What are transfer functions?
• If a transfer function is written in standard form, then the dynamic
properties of the process become apparent
• e.g. time constant of the process and process gains.

𝐾
𝜏𝑠 + 1
• Transfer function models are only directly applicable to processes that
exhibit linear dynamic behaviour
Blending tank example 1
𝑑𝑥𝐴 w1 w2
𝜌𝑉 = 𝑤1 𝑥𝐴1 + 𝑤2 𝑥𝐴2 − 𝑤3 𝑥𝐴 xA1 = f(t) xA2 = 1
𝑑𝑡
At steady state:

0 = 𝑤1 𝑥𝐴1
ҧ + 𝑤2 𝑥𝐴2 − 𝑤3 𝑥𝐴ҧ V
xA w3
Deviation from steady state: xA3
Assumptions:
𝑑𝑥′𝐴
𝜌𝑉 = 𝑤1 𝑥′𝐴1 − 𝑤3 𝑥′𝐴 • Constant Density
𝑑𝑡 𝜌𝐴 ~ 𝜌𝐵
Where:
• Concentration in the tank is the
𝑥′𝐴1 = 𝑥𝐴1 − 𝑥𝐴1
ҧ same as the exit
′ 𝑥𝐴 = 𝑥𝐴3
𝑑𝑥𝐴 𝑑(𝑥𝐴 −𝑥ҧ 𝐴 ) 𝑑𝑥𝐴
𝑥′𝐴 = 𝑥𝐴 − 𝑥𝐴ҧ = =
𝑑𝑡 𝑑𝑡 𝑑𝑡
• Total flowrates are constant
Blending tank example 1
𝑑𝑥′𝐴 w1 w2
𝜌𝑉 = 𝑤1 𝑥′𝐴1 − 𝑤3 𝑥′𝐴 xA1 = f(t) xA2 = 1
𝑑𝑡
Laplace transform:

𝜌𝑉(𝑠𝑋𝐴′ 𝑠 − 𝑥 ′ 0 ) = 𝑤1 𝑋𝐴1
′
𝑠 − 𝑤3 𝑋𝐴′ 𝑠 V
xA w3
𝑋𝐴′ 𝑠 𝑤1 xA3
′ =
𝑋𝐴1 𝑠 𝜌𝑉𝑠 + 𝑤3 Assumptions:
• Constant Density
𝑋𝐴′ 𝑠 𝐾 𝜌𝐴 ~ 𝜌𝐵
′ = = 𝐺(𝑠)
𝑋𝐴1 𝑠 𝜏𝑠 + 1
• Concentration in the tank is the
Where: same as the exit
𝑥𝐴 = 𝑥𝐴3
𝑤1 𝜌𝑉
𝐾= 𝜏=
𝑤3 𝑤3 • Total flowrates are constant
• 𝑥𝐴′ 0 = 0
Properties of transfer functions
• The order of the polynomial in the denominator is the same as the
order of the differential equation.
• This is also known as the characteristic equation

• We can easily determine steady state

• Recall Final value theorem
• Gives us another definition for K
Blending tank example 2
𝑑𝑥′𝐴 w1 w2
𝜌𝑉 = 𝑤1 𝑥′𝐴1 + 𝑤2 𝑥′𝐴2 − 𝑤3 𝑥′𝐴 xA1 = f(t) xA2 = g(t)
𝑑𝑡
Laplace transform:

𝜌𝑉 𝑠𝑋𝐴′ 𝑠 − 𝑥 ′ 0 ′
= 𝑤1 𝑋𝐴1 ′
𝑠 + 𝑤2 𝑋𝐴2 𝑠 − 𝑤3 𝑋𝐴′ 𝑠 V
′ ′ xA w3
𝐾1 𝑋𝐴1𝑠 𝐾2 𝑋𝐴2
𝑠 xA3
𝑋𝐴′ 𝑠 = +
𝜏𝑠 + 1 𝜏𝑠 + 1 Assumptions:
• Constant Density
𝑋𝐴′ 𝑠 𝐾1 𝑋𝐴′ 𝑠 𝐾 𝜌𝐴 ~ 𝜌𝐵
′ = = 𝐺1 (𝑠) ′ = = 𝐺2 (𝑠)
𝑋𝐴1 𝑠 𝜏𝑠 + 1 𝑋𝐴2 𝑠 𝜏𝑠 + 1
• Concentration in the tank is the
Where: same as the exit
𝑥𝐴 = 𝑥𝐴3
𝑤1 𝑤2 𝜌𝑉
𝐾1 = 𝐾2 = 𝜏=
𝑤3 𝑤3 𝑤3 • Total flowrates are constant
• 𝑥𝐴′ 0 = 0
Properties of transfer functions
• When a single process output (O) is influenced by more than one
input(In), the transfer functions are additive.

𝑂 𝑠 = 𝑮𝟏 𝒔 𝐼1 (𝑠) + 𝑮𝟐 𝒔 𝐼2 (𝑠) + ⋯ + 𝑮𝒏 𝒔 𝐼𝑛 (𝑠)

𝐼1 𝑠 𝑮𝟏 𝒔
+
𝑂 𝑠
+
𝐼2 𝑠 𝑮𝟐 𝒔
Properties of transfer functions
• When a process output (O) is a result of a sequence of processes, the
transfer functions are multiplicative.

𝑂 𝑠 = 𝑮𝟏 𝒔 𝑮𝟐 𝒔 𝐼(𝑠)

𝐴 𝑠
𝐼 𝑠 𝑮𝟏 𝒔 𝑮𝟐 𝒔 𝑂 𝑠