Setpoint Tracking Versus Load Recovery

The Core Idea

Setpoint tracking follows a change in the requested target. Load recovery responds to a change acting on the process while the target stays the same. These are different jobs for the same controller.

What you will learn

  • Distinguish a requested target change from a disturbance to the process.

  • Read the setpoint (SP), process variable (PV), and controller output (CV) evidence for each event.

  • Restore the requested temperature without moving SP to conceal an error.

Why it matters

A tune that handles a new target well may behave differently during a disturbance. Both operating situations belong in a useful evaluation.

Same heater, two different jobs

Both experiments start at 25 °C, initially target 50 °C, and use the same proportional-integral (PI) settings: P = 2% output per °C, I = 0.15% output per (°C·s), D = 0. Output is 0–100%; process lag is 12 s with 3 s dead time.

Predict which signal changes first when an operator requests a new target. Then predict which signal changes first when an added cooling load acts on the process while the requested target stays the same.

Request a new target

Run the first heater toward 50 °C. Near 60 simulated seconds, pause it, change Setpoint to 60 °C, and resume. Choose a slower simulator speed if needed, and record the actual time of your change.

Keep the gains fixed and finish the full 150 s. The supplied reference makes the same target change at exactly 60 s; your timing may differ slightly. Compare the SP line, the delayed PV response, and the new output needed to hold the higher temperature.

For another trial, set SP back to 50 °C before Restart. Restart retains the settings you have selected, so check them rather than assuming it restores the original target.

Apply a load with SP unchanged

The second heater keeps SP at 50 °C throughout. At 60 s, an additional cooling load equivalent to 18 °C at equilibrium begins and stays on for the rest of the run. The process, starting temperature, and initial gains match the first copy.

Run for 150 s. Identify when PV starts to depart from its target, how CV responds, and whether PV returns while the cooling load is still present. The target is locked so the check measures recovery to the requested value.

The recovery limit is an average absolute error no greater than 0.5 °C over the final 30 s, with a peak no higher than 52.5 °C during the run. The full run must finish. These temperature limits correspond to 2% and 10% of the original 25 °C step span.

Explain the difference

For setpoint tracking, the operator changes the target and the controller moves the process toward it. For load recovery, the requested target stays fixed and the controller changes output to counter a process disturbance.

On this self-regulating heater, holding a higher temperature or opposing an extra cooling load requires a different sustained CV. Recovery does not mean that CV must return to its old value.

Try I = 0, restart, and compare the ending gap under load. Restore a working PI setting for another full trial. Changing the target to the temperature the heater happens to reach does not satisfy the original operating request.

Hint — Read the target line first

Check whether SP changed at the event. Then follow PV and CV. A PV dip with SP unchanged supports a load-disturbance explanation in this supplied experiment; in a real loop you would also check the sensor and equipment.

Worked interpretation after both runs

At the requested target change, SP rises from 50 to 60 °C. Heating increases and PV follows after the delay. In the load experiment SP remains 50 °C, PV falls, and the controller increases heating to oppose the cooling load.

P = 2, I = 0.15, D = 0 meets the recovery limits. P = 1.5, I = 0.1, D = 0 is another acceptable response. P alone leaves a persistent gap here. Explain your result using temperatures and the CV trace rather than copying one preferred tune.

Check your understanding

For each experiment, record the event, whether SP changed, the first observed PV movement, and the final CV. Explain why a different final CV can be correct even though PV returns to the original target.

Enable JavaScript to run the embedded simulator and complete the interactive exercises.