In many of my previous articles I have highlighted the fact that the vast majority of control practitioners have no real understanding of the practical aspects of regulatory control. They have found that they have been unable to apply the highly theoretical teachings taught to them by the educational institutions they attended, and as a result they have learnt their craft by 'trial and error'. In these articles I have also attached blame to these institutions and have commented they are not teaching the practical side, and in fact appear to have no interest in doing so.
Various people have responded to these articles, and have also expressed their concern about the lack of knowledge displayed by people working in this field. However, in the July issue of SA Instrumentation and Control, Dr Maurice McDowell, writing on this subject, queried if it is the function of educational institutions to teach the practical side.
This has made me think long and hard. It takes me back to the first day of the second year of my studies in electrical engineering at the University of the Witwatersrand, which was the first time we actually started learning truly engineering subjects. Our lecturer began the session by stating that if we thought we had come to university to learn engineering then we had thought wrong. He then commented that what we would learn at university would be to think like engineers. The actual engineering knowledge would be learnt later.
This turned out to be very true. Most of our studies had very little to do with practical engineering.
On reflection I now also think that teaching practical control techniques in an entirely academic environment like a university would be difficult, as one could not immediately go out and practice the learning in an actual plant.
I have very little knowledge about the teachings given at Technikons and other similar institutions where they teach control, but I have always assumed that they would teach a more practical approach, and thought that they would combine much more practice with the theory than do universities. However, in general, I have had just as little interest from them about my ideas on practical control as I have from university academia.
I am certainly aware that very few people in the academic side in control have much field experience, which is possibly why they may be reluctant to talk to me. I do know that for many years it has been impossible for academic institutions to get experienced people back from industry to lecture. My tutor, the late and revered Professor Bozzoli, dean of Electrical Engineering, and subsequently vice chancellor of the University of the Witwatersrand, once told me that there was virtually no chance of a graduate ever returning to teach, once he or she had left university to work in industry. Apart from the salary situation, it is very difficult to go back to the 'ivory tower'. For this reason most lecturers are graduates or post-graduates who had never practiced in a real plant for any extended period.
In his article, Dr McDowell then raised the question as to who should be responsible for training people on the practical aspects on control, and postulated that it is probably the responsibility of industry. This is not a new idea. A new graduate needs a certain amount of practical experience before he can apply for the status of a professional engineer. When I graduated in the early 1960s, a graduate in the United Kingdom had to undergo what was termed a 2-year 'graduate apprenticeship' before he or she could practice professionally as an engineer. (I do not know if this still applies today.)
Although I studied in South Africa, I thought at the time that a graduate apprenticeship would be an excellent idea. Unfortunately the company I went to in the UK, although very large and prosperous, had a very poor graduate apprenticeship training scheme, which was furthermore hampered by problems with trade unions. This meant that the apprentices were not allowed to do anything that was 'proscribed' union work. This effectively meant that you could do nothing except some of the most menial tasks. The local graduate apprentices referred to it as a 'post graduate boredom' and said you just had to sit it out. This was a complete waste of two years of what should be amongst the most fruitful in a young engineer's career. (As it was it actually took me nearly a year to get myself transferred into a permanent position in a department whose work interested me.) It still hurts me to think of all the talent in that company that was wasted for two years. So my own practical learning only came much later.
In my experience very few companies except some of the largest have any formalised practical training for graduates. Therefore, the only way most people can learn the practical skills is by picking them up 'on the job', mostly learning from more experienced practitioners in that field. Now this presents a major problem when it comes to regulatory control, because the really vast majority of people including very experienced individuals do not have knowledge of the practical side, which has never been taught to them in the first place. Their own 'skills' have been learnt mostly by trial and error, and much of that 'learning' has given them instincts which are often wrong.
I have mentioned many times in the past that it is an amazing fact of life to me that in 99,99% of all plants, optimisation of control loops has never been treated as a separate discipline. This is because of the complete state of ignorance of this subject, the general acceptance of the performance of controls over the years, and the fact that very few people are interested enough to evaluate what almost unbelievable paybacks and opportunities could be achieved if the base layer controls in nearly every plant worked properly. Most people do not know how to judge the performance of a control loop. Often the fact that it is in automatic is taken as a sign of satisfactory performance. Therefore, there is little interest, and in most plants optimisation is thought of as 'tuning', which is often left to lower level disciplines, where the people have even less idea on how to go about it.
As there is so little awareness on the need for optimisation, it is unlikely that either industry or educational institutions will ever be interested enough to institute a formal training programme on practical control.
How does one try and raise this awareness? I have been trying to do so for years, with unfortunately very limited success. Many companies have shown an interest, and sent delegates on my courses. Unfortunately even though the trainees themselves usually go back with marked enthusiasm, they very seldom ever get any management support, and pressures are put upon them to perform other purportedly more important tasks.
I have come to the definite conclusion that any optimisation effort is doomed to failure without really enthusiastic support and commitment from both top and middle management.
Two things have occurred in recent years which could spell some hope for raising the awareness. The first of these is that there is a big move afoot worldwide for plants to install advanced control systems. To site an example a year ago, the chairman of one of South Africa's biggest industrial giants stated in his annual financial report that the future of his company lay in the successful implementation of advanced control systems over the next decade.
Now advanced control systems have been around for quite a long time, but it has become apparent to most experienced advanced control engineers that most installed systems have not worked optimally, mainly due to the poor performance of the 'base layer', ie, the regulatory controls. In fact one of the leading advanced control engineers in the South African mining industry recently commented that installation of an advanced control system has generally only resulted in a very expensive method of 'propping up' a poorly optimised base layer.
Many of these advanced control people are now pushing for properly optimised regulatory controls. Again the problem is that they do not seem to be able to get the support of the top management who find it very hard to believe that their regulatory control systems which cost them multimillions, and have been working for years, are not operating properly, and that more money and people must be invested to get them working better.
The second ray of hope which I think may really come into its own, is the fact that a relatively new product generally known as a 'loop performance monitor', is now available in the market. This device can monitor the performance of all the control loops simultaneously in a plant, and can relatively quickly and easily highlight most of those that are not performing properly.
A local manufacturer of a very good loop performance monitoring system, has been installing them on a 60 day trial basis in plants. The results have been absolutely staggering. In every plant where they have done this to date, the results have shown overwhelmingly how badly most loops are performing. Even the biggest sceptic in these plants has been forced to realise that something needs to be done to correct the situation.
Certain personnel in one plant stated that they were aware of these problems, but there was no cure for them because of the nature of the processes. To test this hypothesis, the production manager got me in for a day to investigate several of the worst and most important loops. The result of this was to prove that all of the loops could operate very much better if proper practical control principles were applied.
I must warn that people in plants should not put too much reliance on loop performance monitors to show up all the problems. They can certainly pick up many, but not all, of the problem loops, and quite a few might be missed whose performance could be substantially improved. The reasons for poor loop performances are also generally not shown up. Therefore, the monitors by themselves can not eliminate the need for a proper optimisation exercise to be performed. Loop performance monitors must therefore be regarded as a very important tool in optimisation, but not as the only tool.
The 'half life' of an optimisation exercise is generally in the region of 6-12 months. This is where a loop performance monitor can really come into its own, because once the controls have been properly optimised, the monitor will then immediately flash a warning if any loops' performance deteriorates.
I believe these devices will play a very significant part in firstly convincing people of the need to optimise their controls, and secondly in acting a 'watchdog' for performance degradation once the loops have been optimised.
Once people really understand the benefits of a properly optimised base layer there should be sufficient impetus for companies to ensure that their people do understand the practicalities of control, and put pressure on the relevant educational institutions to teach these skills.
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.
For more information contact Michael Brown Control Engineering, 011 486 0567, 011 646 2385, [email protected], www.controlloop.co.za
| Email: | [email protected] |
| www: | www.controlloop.co.za |
| Articles: | More information and articles about Michael Brown Control Engineering |
© Technews Publishing (Pty) Ltd | All Rights Reserved
printer friendly version