PID Controllers, 2nd Edition

Chapter 7.1 - Control Paradigms: Introduction

So far we have only discussed simple control problems with one control variable and one measured signal. Typical process control systems can be much more complex with many control variables and many measured signals. The bottom-up approach is one way to design such systems. In this procedure the system is built up from simple components. The systems can be implemented in many different ways. Originally it was done by interconnection of separate boxes built of pneumatic or electronic components. Today the systems are typically implemented in distributed control systems consisting of several hierarchically connected computers. The software for the distributed control system is typically constructed so that programming can be done by selecting and interconnecting the components. The key component, the PID controller, has already been discussed in detail. In Section 3.5 we showed that integrator windup could be avoided by introducing nonlinearities in the PID controller. In Chapter 6 it was demonstrated that controllers could be tuned automatically, also that the changes in system behavior could be dealt with by gain scheduling and adaptation. In this chapter, we present some of the other components required to build complex automation systems. We also present some of the key paradigms that guide the construction of complex systems.

A collection of paradigms for control are used to build complex systems from simple components. The components are controllers of the PID type, linear filters, and static nonlinearities. Typical nonlinearities are amplitude and rate limiters and signal selectors. Feedback is an important paradigm. Simple feedback loops are used to keep process variables constant or to make them change in specified ways. (Feedback has been discussed extensively in the previous chapters.) The key problem is to determine the control variables that should be chosen to control given process variables. Another problem is that there may be interaction between different feedback loops. In this chapter we discuss other paradigms for control. Cascade control

07_1-1.jpg

Figure 7.1 Block diagram of a system with cascade control.

is one way to use several measured signals in a feedback loop. (See Section 7.2.) Feedback is reactive in the sense that there must be an error before control actions are taken. Feedforward is another control concept that is proactive because control actions are taken before the disturbance has generated any errors. Feedforward control is discussed in Section 7.3. Model following is a control concept that makes it possible for a system to respond in a specified way to command signals. Section 7.4 presents this paradigm, which also can be combined very effectively with feedback and feedforward. Difficulties may arise when several feedback loops are used. In Section 7.5 we describe some nonlinear elements and some associated paradigms: surge tank control, ratio control, split range control, and selector control. In Sections 7.6 and 7.7 we discuss neural and fuzzy control. These methods can be viewed as special versions of nonlinear control. In Section 7.8 we discuss some difficulties that may arize in interconnected systems. Section 7.9 uses an example to illustrate how the different components and paradigms can be used. Some important observations made in the chapter are summarized in Section 7.10.

UNLIMITED FREE
ACCESS
TO THE WORLD'S BEST IDEAS

SUBMIT
Already a GlobalSpec user? Log in.

This is embarrasing...

An error occurred while processing the form. Please try again in a few minutes.

Customize Your GlobalSpec Experience

Category: Motion Controllers
Finish!
Privacy Policy

This is embarrasing...

An error occurred while processing the form. Please try again in a few minutes.