The history of Proportional-Integral-Derivative (PID) control goes back centuries, but according to the Wikipedia entry, PID Controller, was formalized in 1922.
…we now call PID or three-term control was first developed using theoretical analysis, by Russian American engineer Nicolas Minorsky. Minorsky was researching and designing automatic ship steering for the US Navy and based his analysis on observations of a helmsman. He noted the helmsman steered the ship based not only on the current course error but also on past error, as well as the current rate of change; this was then given a mathematical treatment by Minorsky. His goal was stability, not general control, which simplified the problem significantly. While proportional control provided stability against small disturbances, it was insufficient for dealing with a steady disturbance, notably a stiff gale (due to steady-state error), which required adding the integral term. Finally, the derivative term was added to improve stability and control.
So now, a century later, I wanted to highlight a 5-part series on ControlGlobal.com by Process Automation Hall of Fame member Greg McMillan, The concealed PID revealed.
- The concealed PID revealed, part 1 – ISA committee 5.9 aims to clarify and promote PID capability with technical report
- The concealed PID revealed, part 2 – Five key informational objectives of the ISA 5.9 technical report
- The concealed PID revealed, part 3 – How two-degree-of-freedom controllers both optimize setpoint response and minimize the effect of load disturbances
- The concealed PID revealed, part 4 – Understand PID subtleties to improve and optimize loop and process performance
- The concealed PID revealed, part 5 – An interview view with Yamei Chen and Cheri Haarmeyer on the practical issues of making an ISA Technical Report a reality, and with James Beall about some key benefits of the report
In part 1, Greg noted how process automation legend Greg Shinskey:
…detailed the use of the PID for almost every type of application in the process industry, and showed how the PID is the best controller for handling unmeasured process input (load) disturbances.
Greg explained that with the ISA5.9 Controller Algorithms and Performance report and his numerous books, he sought to:
…keep his [Shinskey’s] legacy alive through my publications and by the automation community via its technical society composing and issuing a comprehensive technical report.
I won’t recap the complete, 5-part series but will share an example of the things in it that you can learn.
PID metrics particularly depend upon the type of process response and are degraded by an increase in dead time. The major types of process responses are self-regulating, integrating and runaway. The open-loop response of a self-regulating process will reach a steady state for a given change in PID output provided there are no disturbances during the test. In a first-order-plus-dead-time (FOPDT) approximation, the response is characterized by a total loop dead time (Θo), an open loop time constant (Τo) and an open loop self-regulating process gain (Ko). If the total loop dead time is much greater than the open-loop time constant, the process is classified as dead-time dominant. If the total-loop dead time is about equal to the open-loop time constant, the process is termed balanced. If the total-loop dead time is much less than the open-loop time constant, the process is termed lag dominant, possibly classified as near-integrating. The open-loop response of a true integrating process is a ramp without a steady state. In a FOPDT approximation, the response is characterized by a total-loop dead time and an open-loop integrating process gain (Ki). The open-loop response of a runaway process is an accelerating process variable (PV) without a steady state. In a FOPDT approximation, the response is characterized by a total-loop dead time, a positive-feedback time constant (Τo’) and an open-loop runaway process gain (Ko’). Often a secondary time constant (Τs) is included in the identification of true integrating and runaway process responses because of its dramatic effect on performance and the benefit of cancellation by rate action. The subscripts “p” denotes a process, “o” denotes open loop, “i” denotes integrating action, “d” denotes derivative action, “v” denotes a valve or VFD, “m” denotes a measurement, and “a” denotes an analyzer.
Click on the links above to read more about the ISA5.9 work and how you can better use the robust and reliable PID control in your industrial processes. For DeltaV distributed control system users, I recommend the whitepaper, Key Features of the DeltaV PID Function Block.