Performance Portrait Method: An Intelligent PID Controller Design Based on a Database of Relevant Systems Behaviors.

Area of Science:

• Control Engineering
• Automation Systems
• Process Control

Background:

• Double integrator plus dead-time (DIPDT) processes are common in industrial applications.
• Designing optimal and robust controllers, such as two degrees of freedom proportional-integral-derivative (2DoF PID) controllers, is crucial for effective process management.
• Existing design methods may not fully leverage available performance data.

Purpose of the Study:

• To develop a computer-supported design method for optimal and robust 2DoF PID controllers for DIPDT models.
• To introduce and validate the Performance Portrait Method (PPM) as an intelligent approach for controller tuning.
• To extend optimality analysis to both parallel and series 2DoF PID controller configurations.

Main Methods:

• Utilizing a database of control tracking and disturbance rejection step responses.
• Assessing performance using speed and shape-related measures of process input/output signals.
• Employing the Performance Portrait Method (PPM) for verifier analysis and iterative design refinement.
• Visualizing the relationship between closed-loop performance and control signal shapes.

Main Results:

• PPM effectively verifies analytical designs and identifies practically relevant controller settings.
• Optimality analysis is extended to series 2DoF PID controllers.
• The method demonstrates efficiency in analyzing parameter effects on optimal processes.
• PPM provides an intelligent, expert-mimicking approach to controller tuning.

Conclusions:

• The Performance Portrait Method (PPM) offers a novel and intelligent strategy for tuning 2DoF PID controllers.
• This approach enhances the design process by leveraging performance data and mimicking expert knowledge.
• The study reveals new insights into PID control design dimensions for DIPDT processes.