Control systems overview

A control system is a device or set of devices used to manage the behavior of other devices or systems.

Process+ is based on linear control systems, i.e. using linear negative feedback to produce a control signal, mathematically based on other variables, to maintain the process within an acceptable operating range.

A control loop is arranged in a system to effectively regulate a critical process variable at an optimal value, called the set point value. Control loops typically include sensors, control algorithms and control actuators. In a control system, a controller is used to manipulate a variable in order to keep the process functioning at the set point value. The difference between the set point value and the current value of the process variable is called the error, which is used to generate an output signal. This output signal is sent to another device, the actuator, to bring the process back to the set point value, via a feedback mechanism. The feedback mechanism will also include some sensing of the error value and the desired results, enabling control systems to adapt to varying process operating circumstances to some extent.

Manipulated variables

A controller can be used to control any process which has the following variables:

  • Process Variable, PV: a measurable output. This is the actual target variable value.

  • Set Point, SP: a known ideal value for the target variable. The optimal target variable value.

  • Manipulated Variable, MV: the manipulated variable will adjust the amount of control exerted on the process variable. an input to the process.

Error values in control systems refer to how far the process variable (PV) is operating from the set point value (SP). Error = SP - PV. The controller attempts to minimize this error by adjusting the manipulated variable (MV).

The controller response is described in terms of responsiveness to the process error i.e. the degree to which the controller overshoots the set point, and the degree of system oscillation.  

PID controllers

A PID controller is a generic control loop feedback mechanism using a well known algorithm. It is very common in industrial control systems. Adding a filter to the controller removes noise signals from the system, and enables better control output.

The PIDF (proportional–integral–derivative-filter controller) controller algorithm involves four separate constant parameters:

  1. Proportional value (P)

  2. Integral value (I)

  3. Derivative value (D)

  4. Filter value (F)

Loop tuning

When a disturbance to the process occurs, the controller needs time to reduce the error signal to zero, and restore the process back to the set point value. By adapting the four parameters in the PIDF controller algorithm, the controller can provide the required control action for this optimal process functioning. This is referred to as loop tuning – optimizing the PIDF values to get more accurate and effective loop control. If the loop is tuned properly, the response time to disturbances will be reduced. If the loop is tuned too aggressively, cycling and overshoot of the set point value can occur.

Control loop performance

  • Well performing loops are defined as when the monitored process variable (PV) varies within the defined process upper and lower control limits.

  • Poorly performing loops describe loops where the monitored PV functions outside of the desired process control limits.

System stability

Processes are in a stable state when there is a zero performance error: the process variable is operating at the desired set point value, so PV = SP.

Making a control change that is too large for a small error value leads to a system overshoot of the set point value. If this occurs repeatedly, and the PV repeatedly overshoots the SP, the PV will oscillate around the set point in either a constant, growing, or decaying sinusoid. If these oscillations increase with time then the system is unstable, whereas if they decrease the system is stable. If the oscillations remain at a constant magnitude the system is marginally stable, although it may still potentially be described as a poorly performing loop if the oscillation values overshoot the process limits.


Related topics:

  

CSense 2023- Last updated: June 24,2025