This post was written by Greg McMillan, industry consultant, author of numerous process control books, 2010 ISA Life Achievement Award recipient and retired Senior Fellow from Solutia Inc. (now Eastman Chemical).
One of the objectives of the ISA Mentor Program is to increase the number of papers and presentations by users to share plant expertise and best practices to advance the automation profession. The paper Effective Use of Key PID Features co-authored with Hector Torres is an excellent example of what is possible from the ISA Mentor Program participants and in the evolving automation profession.
Hector Torres is a senior process and control engineer for Eastman Chemical Company in Tlaxcala, Mexico, with more than 20 years of experience.
The PID controller has evolved over the decades to include a large number of features to address common application opportunities. Many of the features are not effectively used due to lack of a consolidation of functional understanding and best practices. The number of options and parameters can seem overwhelming and the best choices unknown.
The capability of the PID is experiencing a renaissance triggered by the advent of smart wireless instrumentation. A particularly key advancement is an enhancement for wireless that has unexpected benefits for simplifying and stabilizing tuning for loops with large and highly variable update times from at-line and off line analyzers, stopping limit cycles, reducing interactions, and recovering from failures. An emerging capability for smart reset action offers the ability to specify when a PID output comes off an output limit and to prevent overshoot or faltering in an approach to setpoint.
There is also an increasing recognition that a key feature where the controller output cannot change faster than the manipulated variable can enable the user to tune for maximum disturbance rejection and still meet other objectives such as coordination and decoupling of loops. Furthermore the PID can be easily extended to unit operation optimization problems usually reserved for model predictive control.
This paper offers guidance for getting the most out of your PID, extending its functionality from basic control to advanced control by the use of important existing and new features for applications encountered in the process industry. The goal is not only to show the flexibility but also how an understanding of the functionality and synergism of features can result in a simplification of tuning the PID.