0
\$\begingroup\$

Does anyone know of a quote by an eminent control theorist or controls engineer or book author who I could reference who said something along the lines of the following:

The process of control system design begins with understanding the dynamic behaviour of the system that is to be optimized.

I'm sure someone did I just don't know who!

\$\endgroup\$
  • 8
    \$\begingroup\$ I wonder who didn't say that. \$\endgroup\$ – jonk Mar 23 '19 at 21:02
  • \$\begingroup\$ Maybe if I ask a different way: Does anyone know of a methodology or step-wise approach for control system design that would show system identification as an early step in the process? \$\endgroup\$ – Bill Mar 23 '19 at 21:45
  • 3
    \$\begingroup\$ Won’t that be chapter 1, page 1 of any decent text book on the topic... \$\endgroup\$ – Solar Mike Mar 23 '19 at 22:29
  • \$\begingroup\$ "... understanding the dynamic behavior ...": Many classic systems already have a minimal model (eg DC motor). In the first chapters the components are considered to be known. A proper understanding for system identification usually requires a discipline entirely dedicated to this. \$\endgroup\$ – Dirceu Rodrigues Jr Mar 23 '19 at 23:12
  • \$\begingroup\$ Maybe if someone just can give me a well-known practical text-book I could reference. There are quite a lot (I'm not a control engineer). \$\endgroup\$ – Bill Mar 24 '19 at 0:25
1
\$\begingroup\$

I will post quotes here as I come across them:

1.

The most important task confronting the control system analyst is developing a mathematical model of the process of interest. In many situations the essence of the analytical design problem is in the modeling: once that is done the rest of the analysis falls quickly into place.

Control System Design: An Introduction to State-Space Methods, by Bernard Friedland

2.

This 3 minute YouTube video proposes a nice, simple five-step process:

STUDY -> MODEL -> ANALYZE -> DESIGN -> VERIFY

Control System Design Process Diagram

Credit: Jonathan Sprinkle, Aug 30, 2013

3.

The process of designing a control system usually makes many demands of the engineer or engineering team. These demands often emerge in a step by step design procedure as follows: 1. Study the system (plant) to be controlled and obtain initial information about the control objectives. 2. Model the system and simplify the model, if necessary. ...

Multivariable Feedback Control Analysis and design by S. Skogestad and I. Postlethwaite, 2001

4.

The construction of knowledge-driven models starts by analyzing the dominant phenomena at play, making simplifying assumptions and writing material, momentum, and/or energy balances around each part of the system under investigation.

Bonvin et al. Linking Models and Experiments, Ind. Eng. Chem. Res. 2016, 55, 6891−6903

5.

An essential feature of control and optimisation strategies is the availability of mathematical models that accurately describe the steady-state and dynamic characteristics of the process in the whole operating range, including its non-linear behaviour.

Hodouin et al. State of the art and challenges in mineral processing control, Control Engineering Practice 9 (2001) 995–1005

6.

From course notes for GEL-2005 Linear Systems and Control Course, U Laval, André Desbiens, 2019:

Il faut en effet bien connaître comment se comporte le procédé si on désire bien le contrôler.

Which roughly translates as:

You really need to know how the process behaves if you want to control it.

| improve this answer | |
\$\endgroup\$
  • 1
    \$\begingroup\$ You should be careful, this is not entirely true, or at least not thorough. There's a whole segment of data-driven control design which does not require a process model. Even the Ziegler-Nichols PID tables do not strictly require a model for a system, only experimental data. One could argue that any particular design method should be best suited for a specific 'class' of systems, albeit its model unknown. \$\endgroup\$ – Vicente Cunha Apr 13 '19 at 3:50
  • \$\begingroup\$ Yes good point. This quote emphasizes model-building. Whereas I was actually interested in a quote that emphasizes "understanding the dynamic behaviour of the system" as the first step regardless of the eventual control system design (at least for MIMO systems). Still haven't found a good one... \$\endgroup\$ – Bill Apr 13 '19 at 15:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.