Introduction to practical controller tuning: understanding the processes first!

Tuning a feedback control system means adjusting parameters in the controller to achieve robust process control. "Robust" in this context is usually defined as the stability of process variables despite load changes, a fast response to setpoint changes, minimal oscillations, and minimal offset (error between the setpoint and the process variable) over time.

“Robust Control”"It's much easier to define than to achieve. PID (Proportional-Integral-Derivative) control is the most widely used feedback control algorithm in industry. It's important for the engineer and/or instrumentation technician to understand how to tune these controllers effectively and with minimal time investment.".

Different types of processes, which have different dynamic (time-dependent) behaviors, require different levels of proportional, integral, and derivative (PID) control action to achieve a stable and robust response. Therefore, anyone attempting to tune a PID controller must understand the dynamic nature of the process being controlled. For this reason, in this course we will explore the main characteristics of processes before discussing techniques for making practical selections of the P, I, and D parameters.

Process Characteristics

Perhaps the most important rule for tuning a controller is to KNOW the process before attempting to adjust the controller's tuning. Unless we properly understand the nature of the process we are trying to control, we have very little hope of success. In this part of the course, we will dedicate time to explaining the different characteristics of processes and how to identify each one.

Quantitative PID tuning methods (which we'll discuss later) attempt to map process characteristics, thereby yielding suitable PID parameters for the controller. The goal of this section is for you to understand the various process types by observing their responses and performing qualitative analysis. This will help you grasp why different tuning parameters are necessary for each process type, rather than simply following a rigid set of steps to tune a PID controller.

We can classify the response of the processes into three groups: self-regulatory(self-regulating), integrative (integrating) and the unstable (runaway). Each of these process types is defined by its response to a stimulus, impulse, or step (a sudden manual change on an output signal acting on the process, for example, the position of a control valve or the state of some final control element). A process “self-regulating”"responds to a unit step or variation in the opening of a control valve by setting a new, stable PV value. A process"“integrative”responds by constantly varying its PV value (ramping) upwards or downwards at a rate proportional (slope) to the magnitude of the change or step produced by the final control element (control valve). Finally, a process…“runaway”"It responds by varying its PV up or down at a rate that increases over time, leading to complete instability with no way to correct it with the action of the controller.".

Self-regulating processes, integrative y runaway They have different control needs. PID tuning parameters that might work well for controlling a self-regulating process, for example, won't work well for controlling an integrative or runaway process, despite some similar characteristics. If we first identify the characteristics of a process, we can then draw a general conclusion about the necessary P, I, and D values for effective control.

Possibly the best method for testing a process to determine its natural characteristics is to put the controller in manual mode and perform a step change on the controller's output signal. It is critically important that the controller is in manual mode when the process characteristics are being explored. If the controller is left in MODE In automatic mode, the process response we observe to a given setpoint or load change will be partly the natural characteristic of the process itself and partly the corrective action of the controller. The controller's corrective action would therefore interfere with our goal of exploring the characteristics of our process. Conversely, by putting the controller in MANUAL mode, we disable the corrective action (effectively removing its influence by breaking the feedback loop between the process and the controller, and between the controller and the process). In manual mode, the process response we observe for an output (manipulated variable) or load change is PURELY a function of the dynamic nature of the process, which is precisely what we wish to recognize.

Testing or exploring the characteristics of a process with the controller in manual mode is often called an open-loop test, because the feedback loop is "open" and is no longer a complete loop. This test is a fundamental diagnostic technique that we will apply in the following articles.

If you liked the article, COMMENT and SHARE!!!

Want to see more about Controller Tuning? Join our free 100% course.