Feedback Lesson 2
Feedback Lesson 2
Lesson 2: A review of
Laplace Transform
THE LAPLACE TRANSFORM (𝓛) 2. Time Delay
• Used to study the complete response Suppose a function 𝑓(𝑡) is delayed by 𝜆 >
characteristics of feedback system, 0 units of time, 𝑓1 (𝑡) = 𝑓(𝑡 − 𝜆). Its Laplace
including the transient response. transform is:
• Define the Laplace transform of 𝑓(𝑡),
denoted by ℒ[𝑓(𝑥)] = 𝐹(𝑠), as a 𝑓1 (𝑡) = 𝑓(𝑡 − 𝜆)
function of the complex variable 𝑠 = 𝐹1 (𝑠) = 𝑒 −𝑠𝜆 𝐹(𝑠)
𝜎1 + 𝑗𝜔, where:
∞
A time delay of λ corresponds to
𝐹(𝑠) = ∫ 𝑓(𝑡)𝑒 −𝑠𝑡 𝑑𝑡 multiplication of the transform by 𝑒 −𝑠𝜆 .
0−
3. Time Scaling
which uses 0− (that is, a value just
before t = 0) as the lower limit of It is sometimes useful to time-scale
integration. equations of motion.
• f(t) is usually denoted as the time For example, in the control system of a disk
domain, while F(s) is usually denoted as drive, it is meaningful to measure time in
the frequency domain. milliseconds. If the time t is scaled by a
• A function f(t) will have a Laplace factor a, 𝑓1 (𝑡) = 𝑓(𝑎𝑡), then the Laplace
transform if it is of exponential order, transform of the time-scaled signal is:
which means that there exists a real
number σ1 such that: 𝑓1 (𝑡) = 𝑓(𝑎𝑡)
1|Feedback Control
System
The transform of the derivative of a signal Mentioned quantities can be computed
is related to its Laplace transform and its either in the time domain or in the frequency
initial condition as follows: domain.
𝑑𝑓 ∞ ∞
ℒ { } = −𝑓(0− ) + 𝑠𝐹(𝑠) ∫ |𝑦(𝑡)|2 𝑑𝑡 < 1, ∫ |𝑢(𝑡)|2 𝑑𝑡 < 1
𝑑𝑡 0 0
For multiple derivatives: ∞ ∞
∫ 𝑦(𝑡)𝑢(𝑡) 𝑑𝑡 = ∫ 𝑈(𝑗𝜔)𝑌(−𝑗𝜔) 𝑑𝜔
ℒ{𝑓 𝑛 (𝑥)} = 𝑠 𝑛 𝐹(𝑠) − 𝑠 𝑛−1 𝑓(0− ) 0 −∞
− 𝑠 𝑛−2 𝑓 2 (0− ) − ⋯ − 𝑓 𝑛 (0− ) 10. Multiplication by Time
Where n is the number of times an equation
Multiplication by time 𝑓1 (𝑡) = 𝑡𝑓(𝑡)
is differentiated.
corresponds to differentiation in the
6. Integration frequency domain:
The Laplace transform of the signal of a 𝑓1 (𝑡) = 𝑡𝑓(𝑡)
𝑡
time function 𝑓1 (𝑡) = ∫0 𝑓(𝜉)𝑑𝜉 , is given by, 𝑑
𝐹1 (𝑠) = − (𝐹(𝑠))
𝑡 𝑑𝑠
𝑓1 (𝑡) = ∫ 𝑓(𝜉)𝑑𝜉
0 Inverse Laplace
1 The inverse Laplace transform, which
𝐹1 (𝑠) = 𝐹(𝑠) allows us to find f(t) given F(s),
𝑠
7. Convolution
Convolution in the time domain
corresponds to multiplication in the
frequency domain. Assume that ℒ[𝑓1 (𝑡)] =
𝐹1 (𝑠) and ℒ[(𝑓2 𝑡)] = 𝐹2 (𝑠): Where σc is a selected value to the right of
all the singularities of F(s) in the s-plane.
𝑡
ℒ[𝑓1 (𝑡) ∗ 𝑓2 (𝑡)] = ∫ 𝑓1 (𝜏)𝑓2 (𝑡 − 𝜏)𝑑𝜏
0
2|Feedback Control
System