PID Controller Tuning

Prerequisite

None

Overview

This one-day course will cover how to perform effective and time-efficient tuning of single-input single-output PID controllers. The lecture material is supported and reinforced through detailed exercises which are performed using a process simulator and INTUNE PID Tuning Tools software.

This course begins with a review of the fundamentals of process control, moves to a practical treatment of PID algorithms, shows how to identify models for each of your applications and culminates in the use of these models in determining the optimal PID tuning parameters. Along the way, you are exposed to easy-to-use techniques for tuning PID controllers and trained on how to recognize those applications for which PID controller tuning is not necessarily the root cause of substandard process control.

Course Agenda

• The Basics and Language of Process Control
  Process Control is a field of expertise which uses terminology and acronyms which are very specific to control. Concepts such as reverse and directing acting, open-loop vs. closed-loop, and parallel vs. interactive PID control will be reviewed and discussed.
• Setting Goals for Your PID Controllers
  What is most important for your applications? The ability to effectively and quickly reject disturbances? The ability to track setpoint changes as quickly as possible? Minimizing any process variable “overshoot” of the setpoint? An in-depth discussion of how to set your goals and how your goals will impact your controller tuning will be presented.
• The Mathematics Behind PID Control
  What are the P, I, D parameters? How do they work? Are all the PID algorithms offered by the control system vendors the same? If not, how do they differ? The class will go into great detail to investigate the differences between the various “flavors” of PID algorithms on the market today. There will also be a review of the different forms of tuning parameters used by these various algorithms. .
• Identification of Processes in Your Facility / How to Model these Processes
  Is the process you are trying to control an “integrating” process? Or is it a non-integrating process? Does it exhibit a large amount of “deadtime” in its response? A detailed discussion is presented on the attributes of different processes and the techniques used to form a “mathematical model” of these processes.
• Guidelines for Trail-and-Error Tuning of PID Controllers
  If you are unable to perform tests on your process in order to obtain an estimate of its model you may have to resort to the “trial-and-error” method of tuning. Guidelines are presented that will provide a structured framework around trial-and-error tuning giving the user a much greater chance at arriving at “good tuning parameters” in a shorter amount of time when compared to random trial-and-error practices.
• Conversion of Process Models into Optimal PID Tuning Parameters for Controlling the Process
  If you have the ability to perform tests which will facilitate the determination of a mathematical model for your process, then finding the best tuning parameters becomes a straightforward three-step process. A detailed discussion of how to “convert” the model of your process into optimal PID tuning parameters is presented. Also a discussion on how to convert between various PID tuning parameters is presented as well.
• Adaptive Tuning
  The benefits of adaptive tuning and non-intrusive loop diagnostics for different process loops is presented along with an exercise which demonstrates the power and utility of this method of tuning PID controllers.