“PID Tuning? Easier Said than Done” It Doesn’t have to be!

PID Loop Tuning in Five Steps

PID loop tuning may not be rocket science, but it’s not a walk in the park either

In industrial automation to control process variation and disturbance in a system, the most common method is PID (Proportional integral derivative) controller.

What is the function of a PID Controller?

A PID controller calculates an error value as the difference between a measured Process Variable (PV) and a desired Set Point (SP). The controller attempts to minimize the error by adjusting the process through the use of a Manipulated Variable (MV).

To optimize the process, we need to tweak the PID manipulated variables. The PID controller algorithm includes three separate constant parameters, the proportional, the integral and derivative values, denoted by P, I, and D.

These values can be interpreted in terms of time:

P depends on the present error,

I on the accumulation of past errors, and

D is a prediction of future errors, based on the current rate of change.

Step One – Discuss

The first basic principal to loop tuning is to understand the process dynamics, its behavior and safety limitations. For a Control Engineer, a detailed discussion with the process and operations team about the ‘what if’ scenarios to identify limits and constraints of the process is necessary.

Step Two – Perform a Step Test

The second step involves loop optimization by performing a Step Test. But before you do that, an open loop test needs to be performed by making changes in Output and monitoring its behavior on Process Variable (PV) irrespective of Set Point (SP). Your next step should include closing the loop and making a step change in Set Point (SP) which should be more than the process noise to see its response clearly. For later behavioral analysis, 6 to 10 step tests need to be recorded for model identification.

Step Three – Open Loop Tuning

The third step includes open loop tuning. For open loop tuning it is necessary to change the loop mode to manual. Open loop tuning helps identify the process or instrument issues like valve parsing, logic error or design error and process limitation. Record your observations in a report and present it to process team for rectification prior to closed loop tuning. 

Step Four – Closed Loop Tuning

The fourth step entails closed loop tuning. For closed loop tuning, you will need to change the loop mode to auto. Closed loop tuning helps identify the PID controller behavior with respect to the current parameters. The default parameters for different types of loop are listed below.

Process Description	Loop Behavior	Proportional	Integral	Derivative
Flow	Fast Loop	Little	More	Not needed
Pressure	Fast Loop	Little	More	Not needed
Temperature	Slow Loop	More	Some	Some
Level	Integrating	More	Little	Not needed
Consistently changing loop	Noisy	Low	Most	Off

Step Five – Trial & Error

The fifth step five is by far the most critical. It includes optimizing the loop by using the trial and error method. The trial and error method is the most commonly used technique to determine initial tuning constants however, further adjustments may be necessary. The biggest advantage of a trial and error method is that it’s simple, logical and deals with all component in the process. Having said that, there are a few disadvantages associated with this method. For instance, it can be time consuming and requires patience and process understanding.

Mentioned below are some of the best practices when using the trial and error method.

Trial and Error Method – Best Practices

Proportional Action	Start with a Low ValueIncrease graduallyReduce if the PV oscillates excessively, if MV has large over shoot or if you see proportional kick offThe important part is the proportional effect both integral and derivative
Integral Action	Start with a High value (low integral action)Increase the integral action if the approach set point is slowExcessive integral can also cause oscillationIntegral time should be > dead time +1/3 lag timeReduce gain when decreasing integral time constantUse more integral action and less proportional action to stop kick back
Derivative Action	Do not use it all the timeUse it when you have dead time and no noiseIntroduce it slowly and carefullyInitially Derivative is (dead time/4)Do not use too much as it can lead to instability if over used, max should be TD=TI/4 or TI/10

In a nutshell, the art of a successful PID loop tuning is to adjust Output to move the Process Variable (PV) as quickly as possible to the SetPoint (SP, responsive). Next, minimize overshoot and hold the Process Variable (PV) steady at the Set Point (SP) without excessive Out Put changes(stable).

PID loop tuning may not be rocket science, but it’s not a walk in the park either. All it needs is some patience and experience.