Why should PID oscillate?
Have you ever thought about the typical oscillations that the PID algorithm arouses when controlling the process variable to the setpoint?
Are the oscillations really necessary from a physical point of view?
Empirical?
The point is that the PID algorithm is a hundred year old empirical formula. In 1930 Ziegler and Nichols tried to find the fundamental working principle of the interaction between the formula and the process/system. Instead they came up with a trial and error method to configure the formula.
It is plausible to regard the PID algorithm as one of the key successes of the industrial revolution. At the moment, 30,000 aircrafts are controlled by the formula as well as 60,000 power plants, 1 billion cars and 2 billion room thermostats. At this time of day, there are over 50 billion PID formulas being recalculated every 100 ms every time.
Inconvenience
Despite the hundred-year success, an uneasy feeling remains.
Why would it be necessary to correct a process governed by simple physics in an oscillatory way?
Research
Recent research shows that there is at least one mathematical- physics inconsistency present in the well-known PID algorithm. This could explain some ambiguities regarding the configuration and the use of the feedback PID control algorithm.
50 billion
The idea of posting some views about PID feedback control was to see if there are fellow engineers who also have doubts about the correctness/validity of the PID algorithm. It is an important topic, as PID is by far the most commonly used control algorithm. As mentioned before, at this time of day, there are over 50 billion PID formulas being recalculated every 100 ms every time.
Efficient helmsman
Nicolas Minorsky (born 1885) is one of the pioneers of the PID-algorithm. He observed and researched manual steering of ships. In 1922 Minorsky wrote:
"An efficient helmsman keeps the ship accurately on her course by exerting a properly timed meeting and eming action on the rudder"
Oscillating?
It is questionable that the PID algorithm represents an efficient helmsman who keeps course by oscillating around the setpoint. Even when the helmsman has little information about his ship and the external conditions (disturbances), it is doubtful whether the oscillations around the set point are necessary from the point of view of physics?
The oscillatory behavior of the interaction of the PID algorithm and the process variable of the process/system is largely caused by the integrator term. In this respect, the oscillating behavior can be explained. It is also clear that the integrator is an important term within the PID algorithm that is responsible for the process variable eventually becoming equal to the setpoint.
Inconvenience
Despite the hundred-year success, an uneasy feeling remains.
Why would it be necessary to correct a process governed by simple physics in an oscillatory way?
Research
Recent research shows that there is at least one mathematical- physics inconsistency present in the well-known PID algorithm. This could explain some ambiguities regarding the configuration and the performance of the feedback PID control algorithm.
For now, I want to explain that the said research unequivocally demonstrates a flaw in the PID algorithm: a flaw in terms of an incorrect interactive behavior of the algorithm (combined with its actuator) on the process/system it controls.
Next
The following LinkedIn article explains why a low-pass filter should not be used with PID feedback control, in fact not with every feedback control.
LinkedIn groups
In the past few weeks we have posted some substantive posts in various LinkedIn groups of control engineering experts. Based on more than 200,000 reads, more than a 1.200 likes and over 150 comments, it can be assumed that control engineering experts consider it not impossible that the PID algorithm contains an inconsistency.
The configuration problems, the dimensionless coefficients and the questionable control performance (oscillations) of the PID algorithm are decisive here. In none of the responses of the control engineering experts are the stated two problems of PID control contradicted. On the contrary.
This article was published in two subsequent posts in the following groups:
Wij maken gebouwen slimmer, zuiniger en veilig | EPBD - GACS - Energie effici?ntie - systeemarchitectuur - cybersecurity
2 年Hi Casper, from a practical point of view a system does not have to oscillate to reach its setpoint. It depends on the requirements: if a quick steady state is the goal: a PID controller with acceptable overshoot (lower than 20%) is the goal. If no overshoot is allowed an aperiodic approach is also feasible (gain-factor < 1) A PID-controller in a feedback controlloop inherently is oscillating on its system frequency.
Working on Advanved Refrigeration Cycle Optimization
2 年As mentioned, this post attracted a lot of attention in various discussion groups on LinkedIn. In recent weeks we have posted substantive posts in various LinkedIn groups of experts in the field of control engineering. Including the information in this LinkedIn article. The posts have been read more than 100,000 times, more than a thousand likes and more than a hundred control engineering experts have commented substantively and shared their experiences with PID feedback control. Control engineering is a highly developed field in which engineers, mathematicians, computer experts and electronics experts have developed advanced knowledge and technologies. Control technology plays a much bigger role than many people know. That's because they aren't big eye-catching machines that move, make noise or spread smoke. High-tech control technology is hidden deep in a cupboard somewhere in the corner of a factory. The reason why control technology plays a major role is because these systems ensure the correct, efficient and safe operation of all large, eye-catching machines that move, make noise and/or spread smoke.
Sr. Service Engineer
2 年Thanks for sharing