How to Tune PID Controllers

There are several approaches to tuning PID controllers:

• Manual, using your knowledge of control theory
• Rule-based, relying on established methods
• Automatically, using software

Using Manual and Rule-Based Methods for PID Tuning

Manual tuning methods such as using the open loop transfer function, rely on your knowledge of control theory. This approach tends to be iterative and time-consuming, and if used on hardware,  can risk damage. Rule-based approaches are also manual, but instead use the work of people with proven methods such as Ziegler-Nichols and Cohen-Coon. A limitation of rule-based methods is that they do not support certain types of plant models, such as unstable plants, high-order plants, or plants with little or no time delay.

Automatically Tuning PID Controller Gains

You can automatically tune PID controllers using software tools to achieve the optimal system design and to meet design requirements, even for plant models that traditional rule-based methods cannot handle well.

An automated PID tuning workflow involves: 

• Identifying the plant model from input-output test data
• Modeling PID controllers (for example, in MATLAB^® using PID objects or in Simulink^® using PID Controller blocks)
• Automatically tuning PID controller gains and fine-tuning your design interactively
• Tuning multiple controllers in batch mode
• Tuning single-input single-output PID controllers as well as multiloop PID controller architectures

To learn more about automated PID tuning, see Control System Toolbox™ for use with MATLAB and Simulink.