PID Control loop parameters - Papermachine Automation

Control loop tuning

PID Control

PID control refers to the control system that alters the process outputs to bring the measured valued closer to the setpoint value.

The computer uses an error value (the difference between the setpoint and the measured value) to base the calculations on. PID is comprised of 3 parts, the proportionality value, the integral action and the derivate action. The majority of control systems tend to use PI control because of the limitations of the D action.

The Proportional Action

The proportional term produces an output that is proportional to the error value. The proportional response to the error value can be multiplied by the proportional gain constant (Value P on the control system).

A high proportional gain constant (Pk) relates to a larger change in the output to correct the error value. If the proportional gain constant value is too high it can make the control loop unstable. Likewise, if the value is too low the control action can be too small and results in a small output when responding to a higher input. This can lead to a less sensitive controller.

The integral action

The contribution to the PID control from the integral term is proportional to both the magnitude (highest value) of the error and the duration of the error. Plainly speaking the Integral value affects the time it takes for the proportional value to meet the set point by reducing the error value to 0. The controller works by adjusting the repeats per minute value. The bigger the integral action the quicker the proportional value meets the set point. By increasing the repeats per minute value the faster/ bigger the integral action is.

.

The derivative action

The derivative action looks at the slope of the error value and tried to predict the measured value. This allows you to have higher P and me values while also keeping the system stable. The value of the derivative action describes how far in the future it should look so 20 would mean it looks 20 seconds into the future.

The issue with using the derivative action is if there is noise on the measured value (small spikes) this confuses the algorithm used and increases the effect of the Derivative value leading to an unstable control system. Most control systems do away with the Derivative action for this reason. The derivative function looks at the “steepness” of the curve, having noise on the MV leads to high steepness but no actual chance this is why the derivative action breaks down.

The PID control display page can be accessed through the face plates on the DCs system but cannot be altered. Fine tuning of the PID controls can be difficult. Fine tuning of the controls should dampen the oscillations created by the output so that the MV matches that of the SP quickly and stabilizes.

For more information of the devices that can be used with PID controllers within the papermaking industry take a look at this; Process Level Control - Paper machine Automation