Closed loop control

From Citizendium
Revision as of 21:15, 14 September 2007 by imported>Hendra I. Nurdin (Stub for closed loop control)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

In control engineering, closed loop control is a control system configuration in which the controller has access to signals which contain information about the current "state" of the plant (the object or system to be controlled) during the time that the controller is in operation. Here state roughly refers to a collection of dynamical variables of the plant which determine the plant's future evolution/trajectory given its future inputs.

In closed loop control, the control law implemented by the controller at any time will depend on values of some dynamical variables of the plant up to the time . In practice, this is facilitated by feeding back some signals from the plant into the controller by interconnecting some ports of the plant to some ports of the controller. For this reason, closed loop control is also referred to as feedback control. The main advantage of a closed loop control configuration over an open loop one is that, for an appropriately designed controller, it can offer robust performance under the presence of uncertainties regarding the plant. For example, in practice it impossible to know the plant perfectly, hence all mathematical models for the plant are ultimately approximations of the true plant and there will inevitably be modelling/approximation errors. Such modelling errors constitute one type of uncertainty which are encountered in control engineering practice. In a closed loop configuration it may be possible to design a controller which can give acceptable control performance despite imperfect knowledge of the plant to be controlled, assuming that the control engineer designing the controller has a reasonabe idea about the most significant type of uncertainties present (for example, based on empirical data collected from the plant). Naturally, closed loop control requires that there are plant signals that can be sensed and usefully fedback to the controller. This may not always be the case for arbitrary plants and in such cases it is necessary to resort to an open loop control configuration.

See also

Open loop control