Glossary of Control Engineering Terms - S

Sampling Period: In a discrete time control system, this is the time between successive instantaneous samples of the process variable. For a digital controller the process must be sampled, a new output calculated and that value sent to the actuator all within the sampling period. Ideally, the sample period should be set between 4 and 10 times faster than the process time constant.

Sampling Rate: The frequency with which samples of a physical variable, such as sound are taken. The higher the sampling rate, the more closely the digitized result resembles the original. This is the inverse of the Sampling Period.

Saturation: When an actuator hits its magnitude or rate of change limits. A controller with integral action needs to be aware (anti-windup) that the actuator is saturated otherwise it will keep integrating.

Scan Time: The time to completely execute a PLC program once, including I/O update. See also Sampling Period.

SCADA (Supervisory Control and Data Acquisition): The level of applications that monitor and control devices such as programmable controllers. These systems are usually PC or workstation based.

Second-Order System: A linear dynamic system that has two poles, usually characterised by a gain, a damping ratio and a natural frequency.

Self-Regulating Processes: Most processes are self-regulating, where a change in the actuator causes the process variable to move and then settle (when in manual/open loop). Integrating loops are non-self-regulating.

Self-Tuning Control: A class of adaptive control, where a parameter estimator is used to track changing parameter in the model that represents the plant and thus recalculate the controller gains.

Sensor: In a closed-loop control system, this is the component that translates the physical process variable into a measurement (the measured variable) that is used within the controller.

Set-Point: The set-point is where you would like the process variable to be. For example, the room you are in now has a setpoint of about 70 degrees. The desired temperature you set on the thermostat is the setpoint. Also called reference signal.

Servo: See Tracking

SISO: A system with one input signal and one output signal (single-input single-output). See MIMO.

Smart Instrument: Smart instrumentation is any field device that, in addition to transmitting the measurement variable or receiving the control variable, also transmits other process information and/or diagnostic information about the device. Many smart instruments also allow the control system to re-programme the device (such as re-ranging it) in situ with minimal impact on the process.

Split Range Control: A control technique where the output of a controller is divided into 2 or more signals that are applied to different control devices. There are several reasons for splitting the signals: there may be a fine control action device and a coarse control action device one control device may be for lower rates and the other for higher rates there maybe parallel streams. Traditionally the signal splitting has been performed in the field by the setting up of the positioner. Today with cheaper controllers split range should be performed in the control system because it gives more flexibility and provides the operator with a better view of the plant status.

Stability: A stable system when perturbed from an equilibrium state will tend to return to that state. An unstable system when perturbed from equilibrium will mover further away at in increasing rate, or for a nonlinear system may move to another equilibrium state. For linear systems stability is checked by the eigenvalues of the state matrix or the poles of the transfer function.

State Estimator: Another name for an observer. The Kalman filter is a state estimator.

State-Space: In a state-space system it is assumed that the complete characteristics of the dynamic system can be represented by an nth-order state vector at every instant. For linear systems the notation is common where the system is a matrix of 1st order differential equations, is the state vector, is the system output and the system input.

Statics: This is the steady-state component of a system response, i.e. once all dynamic behaviour has decayed, and is essentially the gain between the inputs and outputs. In a physical sense this is determined by things like actuator sizing, motor coefficients and thermal transfer coefficients.

Steady-State: The state reached once all of the dynamics of a system response have decayed. See also statics.

Steady-State Gain: The steady-state relationship between a system's input and output is represented by the gain of the system, which is a scalar for a SISO system or a matrix for a MIMO system.

Supervisory Process Control (SPC): The use of a single processing entity to generate setpoints for a control system. Often the generation is based on data from many sources, for example the control system itself, laboratory results, and production schedules. Note that the entity implementing SPC can have a backup in case of failure.

System Integration: The process of connecting systems, devices, and programs together in a common architecture so as to share and exchange data.