Control Loop Case History 120

Control loop optimisation: theory versus practice?

Virtually all feedback control courses worldwide are presented in an extremely theoretical way. This is understandable, as feedback control theory was largely developed back in the early 1900’s by some of the world’s leading mathematicians, including Bode, Nyquist, and the father of PID, Nichols. Mathematicians, as most of us well know, don’t come from planet Earth, but from the extraordinary planet Mathematica, where the denizens speak the strange language of Mathematics which is extremely hard for simple mortals like us to understand.
The vast majority of control people after graduating from their places of learning, arrive in a plant to find that apart from a few very rare individuals, no one involved in the feedback control field use any mathematics. This is because there are no mathematical models of process transfer functions, so people cannot use any of the fine things they have learnt. They then resort to learning to “fly by the seat of their pants”, and start on the path of PID “knob twiddling”, hoping that this will solve all the control problems.
This is borne out by study undertaken by a control journalist some years ago in the USA. He investigated tuning methods in over 135 plants in both Northern America and Canada, and reported that he found that 98% of all tuning was done in plants by using either the WAG or SWAG methods. This is his definition of those methods:
WAG: Wild Arse Guess
SWAG: Scientific Wild Arse Guess
Another common definition of most tuning methods is “Trial and Error”, but in the case of one of South Africa’s leading paper plants, it is termed “Trial and Horror”.
Basically what happens is that people play around with the controls and over the years start developing their own idea and feelings as to how to go about optimisation. I find that many of these ideas are completely wrong. In discussions on control on a professional control social network, I frequently come across contributions by people who are supposed to be experts on the subject. Many of these are complete nonsense, which are obviously based on empirical knowledge or feelings, and are completely wrong. Also when giving courses I find that many of the more mature and experienced practitioners of control initially find difficulty in accepting many of the things I teach, as they go against concepts they have accepted for years as correct.
Some of the topics that people have completely the wrong idea about include:
1.When, and how to use the D (derivative) term in the PID controller. There is more nonsense talked about this than virtually any other thing.
2.Understanding how to properly set-up PID controller in PLC’s. Typically they are not working correctly in 85% of plants I work in. In several cases I have encountered, they are literally not working at all, and the people on site were not even aware of it.
3.Understanding how controllers work and what the various options are for. This can cost plants a huge amount of money; typically £14 million per annum in under-production in one pharmaceutical plant in the UK.
4.How to scientifically tune PID controllers.
5.How to analyse and trouble-shoot control loops. About 50% of all control loops I encounter have some sort of fault or problem in them, which the plant people generally are not aware of.
6.The widely held belief across all disciplines in a plant that all control problems can be solved by tuning. In reality tuning is the very last procedure that should be performed when optimising control loops.
7.The almost universal belief that all tuning should be set for quarter amplitude (sometimes called quarter wave damping) response to setpoint step changes. This is taught in all theoretical control institutes. In reality it is not good for practical reasons. Amongst other things it means that the control valve goes through approximately 8 reversals every time the controller tries to correct an error, which is ridiculous. It is also not robust enough for real life control. (However in reality I find about 85% of all control loops are tuned probably anything from 20 to 1,000 times slower than quarter wave response. This is because very few people have any real idea at all on how to tune loops.)
8.Understanding the differences between the two classes of processes, namely self-regulating (e.g. flow control, and integrating (e.g. level control). They behave differently, and are also tuned completely differently.
9.Understanding even basic process dynamics, and how to tune for the different types of dynamics.
10.The belief by many process engineers and specialists that anything can be controlled perfectly by a PID controller. There are often limitations to control, which they are not aware of.
11.The belief by many process engineers and specialists that control optimisation is a very simple process, and many also believe that it can be carried out by low level disciplines, and that control specialists are not necessary.
12.The belief by many process engineers and specialists that they themselves have no need for knowledge of control, and it can all be handled by their C&I people. In fact it is almost essential that they too should have an excellent understanding of control as they are the people who have the process knowledge and should be able to give guidance to the C&I people (who often do not really understand the process), as to the purpose of the control and how it should operate. They should also understand the limitations of feedback control.
13.Many plants, particularly in the mining industry, are designed by process specialists who have no knowledge of, or interest in control. The controls appear to be added as an afterthought. As a result control strategies are sometimes incorrect, or cannot work properly. Typical examples of this in the mining extraction business are sumps with insufficient retention times so they cannot smooth out surges, and flotation tank lines without an upstream surge tank so that a constant flow of product cannot be fed to the line.
14.The belief that if a loop cycles it must be due to bad tuning. In actual fact there are about 8 different things apart from tuning which can create cycling in a loop.
The people who generally are the most excited after completing our courses on practical control are firstly Control and Instrumentation specialists with years of experience. They usually come out of the courses wishing that they had acquired that knowledge years ago. The second group who find it extremely beneficial are process engineers and specialists like metallurgists. One Chemical Engineer (who had graduated in with a First in control), and is now a manager of a paper machine in large fine paper plant said to me at the end of the course, that she had had no previous idea whatsoever of the intricacies of practical process control and what was involved, and stated that she would push for all their process managers, engineers, and specialists to attend the course, because she now realised how important it was for them to have that knowledge.
For an example of problems in this article, I will show another one of a loop in a continuous cycle which was not caused by bad tuning. (Case History 118 dealt with several others.)
The loop in question was a level control where the requirement was to keep the level on setpoint, and to deal with load upsets as quickly as possible. The problem being experienced in the plant was that the loop was always cycling, and of course everyone had been trying for years to tune it to stabilise it.

Fig. 1

Figure 1 shows the open loop test. There are two things of interest in the test. Firstly there is a deadtime of about a minute before the level starts responding to a change in the output of the controller. This is really excessive, and we were not able to find out what was causing this. (Normally one would expect to encounter a deadtime of a few seconds maximum on such a process.) This makes the process very difficult to control, as deadtime means that one must slow the controller down to try and avoid instability. Secondly a small change on the output caused the level to change very quickly, as the tank had a very small retention time, which means relatively large process gain. This too makes it difficult to control.
Both excessive deadtime and large process gain means that the controller has to be tuned slowly to avoid instability. Thus only a very low gain can be used in the controller. However even with such a slow tune the loop still continued to cycle. One of the first thing people are taught in our practical optimisation course is that if one has hysteresis on the valve in an integrating process, and if you are using P+I control, then a continuous cycle WILL result.

Fig. 2

Figure 2 shows the closed loop test with the cycle going on. It is a classical case of an integrating loop hysteresis cycle! The cycle was probably being intensified because of the deadtime, but unfortunately short of removing the integral term from the controller, tuning cannot stop the cycle. This cannot be done in this case as the low proportional gain in the controller could result in large offsets from setpoint. The only thing that can be done here is to fix the valve, and possibly also investigate why the deadtime is so excessive. Much better and faster control could be obtained if the deadtime was to be reduced.
This example is a good example of how important it is to understand the practicalities of control, as these sorts of things are not discussed in theoretical courses where everything is supposed to be perfect. Unfortunately as we all know it is definitely not a perfect world, particularly when it comes to process control.

Index to articles

Michael Brown is a specialist in control loop optimisation, with many years of experience in process control instrumentation. His main activities are consulting, and teaching practical control loop analysis and optimisation. He gives training courses which can be held in clients' plants, where students can have the added benefit of practising on live loops. His work takes him to plants all over South Africa, and also to other countries. He can be contacted at:
Tel (011) 486-0567
Fax (011) 646-2385
E-Mail:  michael.brown@mweb.co.za