Controller Tuning Methods:
Ziegler-Nichols and Cohen-Coon
Practical Techniques for Optimizing Control System Performance
Presented by: Pranjal Gupta
Content
• Introduction to Controller Tuning
• Types of Controllers
• Ziegler-Nichols Tuning Method
• Ziegler-Nichols: Advantages and Limitations
• Cohen-Coon Tuning Method
• Cohen-Coon: Advantages and Limitations
• Comparison of Tuning Methods
Introduction to Controller Tuning
What is Controller Tuning?
Control tuning is the process of adjusting the parameters of a controller, such
as a Proportional-Integral-Derivative (PID) controller, to achieve optimal
performance of a control system. The goal of tuning is to ensure that the system
responds to changes and disturbances in a stable, efficient, and accurate
manner.
Objectives:
• Achieve fast response without excessive overshooting.
• Minimize steady-state error.
• Maintain system stability under varying conditions.
Types of Controllers
Parameters to Tune:
• Proportional (P): Determines the reaction strength to the error.
• Integral (I): Eliminates steady-state errors by considering the cumulative error
over time.
• Derivative (D): Predicts system behavior and reduces overshooting by considering
the rate of error change.
Combination Types:
• PI, PD, and PID controllers.
Ziegler-Nichols Tuning Method
• Think of it as the "trial and error" method.
• It involves gradually increasing the controller's gain until the system
starts to oscillate (like a swing going back and forth).
• Based on the oscillation frequency, the method provides a set of
initial tuning parameters.
• It's simple but might not always give the optimal results.
Ziegler-Nichols Tuning Method
Steps-:
1- Remove the integral and derivative modes of the controller by setting 𝑅 →
0 𝑎𝑛𝑑 𝐼 → ∞. Then the controller becomes proportional.
1
𝐺𝑐 𝑠 = 𝐾𝑐 1 + 𝑅𝑆 + 𝐺𝑐 𝑠 = 𝐾𝑐
𝐼𝑆
2- Select a small value of the proportional gain Kc.
3- Disturb the system with a step change.
4- Observe the transient response of the output variable.
5 - If the system decays , then select a higher value of Kc and again observe the
response of the system
6 - Continue increase the gain Kc in small steps until the response first exhibits
sustained oscillation.
Tc
0 t
7 - The value of the gain Kc and period of oscillation Tc that correspond to the
sustained oscillation are the ultimate gain (Kcmax) and ultimate period of
oscillation Tc.
8 - From The value of the gain Kc and period of oscillation Tc we can predict
the optimum values of the controller parameters using Ziegler-Nichols rules
as shown in the Table below.
Ziegler-Nichols: Advantages and Limitations
Advantages:
• Simple and quick.
• Suitable for many industrial systems.
Limitations:
• May not work well for systems with long dead time.
• Can result in oscillatory behavior.
Cohen-Coon Tuning Method
• This method is a bit more advanced.
• It requires analyzing the system's response to a step change (a sudden input).
• Based on the system's reaction time and other characteristics, the method
calculates the optimal tuning parameters.
• It often provides better results than Ziegler-Nichols, but it requires more analysis.
Cohen-Coon Tuning Method
This method is also called reaction curve method or open loop method in
which the control action is removed from the controller by placing it in
manual mode.
Cohen – Coon approach
1- Switch the controller to manual mode. Split the controller from the closed
loop.
2- Introduce a step change in the controller output M(s) that goes to the
valve and record the transient response (B)
3- The response of the system (including the valve, the process and the
measuring element ) is called the process reaction curve. This response will
appear as S-shape
y
0
Time
4- Draw the horizontal asymptote to the final response KA
y
KA
0
t
5- locate the inflection point on the response curve f , then draw a tangent to the
curve from the inflection point (line ab in the figure below)
y
b
KA
0
a
t
6- Draw a vertical line from (b) to x-axis and locate point (c) as shown in figure
below.
7- From the plot, calculate both; 𝜏 𝑎𝑛𝑑 𝜏𝐷 as shown in Figure below.
y
b
KA
0
0 a
c
𝜏𝐷 t
𝜏
8- From values of 𝜏 and 𝜏𝐷 we can estimate the optimum values of the controller
parameters ( Kc, R and I) as shown in the table below.
Cohen-Coon: Advantages and Limitations
Advantages:
• Provides better control for processes with dead time.
• Reduces overshoot compared to Ziegler-Nichols.
Limitations:
• Requires accurate open-loop testing.
• More complex than Ziegler-Nichols.
Comparison of Tuning Methods
Feature Ziegler-Nichols Cohen-Coon
Ease of Use Simple Moderate
System Oscillation Higher Lower
Dead Time Handling Poor Better
Applications General systems Dead time systems
Thank You
