Ask Dr Bob: What is closed-loop control?

Video Transcript:

Hello, I’m Bob Rice. I’m the Head of Engineering for Control Station. And in this video, we’re going to be covering what closed-loop control is. To do that, I’m going to go ahead and draw three tanks onto the screen over here. And we’re going to take you through three different ways to control the liquid level inside of these tanks.

So these are three identical tanks, and we’re going to start at the top tank here, we’re going to go ahead and draw an outflow, and we’re going to say we have a control valve that’s feeding liquid into this tank. So we’re gonna have a little control valve right here. And for our first example, we’re going to do manual control, or what is known as open loop control. In which case, there’s actually an operator or a person that’s sitting up there watching this tank, there’s no level indication on there, there’s actually probably somebody sitting up in scaffolding, looking at the tank to say, hey, is the liquid level too high or too low, if the liquid level is too low, they’re going to go ahead and manually open and close this valve. Okay? If the liquid level is too high, they will make the adjustments, so they are manually adjusting that valve position themselves. Okay? That is known as manual control or open loop control. All right, manual or open loop operator driving the valve.

Next, we have the same scenario, we’ve got liquid level flowing out of the tank, we have liquid flowing into the tank, in which case, we have a control valve that’s here. But instead of an operator physically inspecting the tank, we actually have a level indicator on it. So we have a level indicator on here. That is measuring a digital signal that tells us how much level is inside that tank. Okay, this is being fed, then to a level controller, which is a computer algorithm, an equation that we’ll talk about a little bit later called the PID equation, that is actually taking a look at this level indication and then adjusting the valve position based off that equation. Okay, so you feed into this some sort of set point that you’re trying to achieve. This measures the level, the difference between the set point and level or set point and measurement is the error, the deviation, it’s going to look at that deviation and make a valve position change to be able to maintain that this is known as automatic. Automatic or closed-loop control.

What is this one in the middle? Well, we’re going to take the same tank. And we’re going to introduce the concept of mechanical control. This is going to be a very simple example, especially for level control. You can have mechanical controls for all types of systems. The level control is kind of unique. You can actually have a control pipe coming in. And you’ve got some sort of float valve that’s on it. And you’ve got some sort of system with a float down below. And then, as that float rises, it can mechanically adjust something in that control valve to close the valve. So as the level starts to rise, the valve starts to close, chopping off the flow of liquid, liquid stops going in right level starts to drop, the valve opens, and the liquid level starts to drop, more liquid starts to flow in, because that valve opens and you kind of maintain some sort of kind of steady state type of level in here. Now, this is known as mechanical control. Okay, and this is going to be like a float system. This would be the same type of thing that you have in your toilet.

Now, which one of these is best? Well, it depends on the application that you’re trying to control, right? Manual control requires somebody to actually physically look at the system and make valve adjustments now for very slow systems, where you may not have a lot of liquid level flowing out very quickly. And you’ve got days between adjustments. Not too bad. Notice there’s no level indication on here. You could put a level indicator and then have some operator out sitting in the control room looking at this and making the adjustments that way, but they’re still manually intervening with it. Okay. Automatic control has an equation on here that’s looking at the level indication and is continuously making this adjustment. Okay. So an operator manually adjusting is a lot of interventions and a lot of labor to be able to control the liquid level. Down here, we have a controller that’s actually making those adjustments for us. The mechanical is very simple. It was very, very cheap. cheap, but it’s not ultra-precise. It’s not very accurate. In fact, we don’t actually know what the liquid level of the tank is. At any point in time, we just have to assume that the mechanical system is doing this, right is adjusting, opening, and closing the valve. Now your toilets at home all use this type of mechanical control. Why, it’s cheap, right? You don’t want to hire somebody to look at your toilet all the time to see how much liquid level is there and open and close the valve. It doesn’t make sense. Just like it doesn’t make sense to put a control PLC or DCS on there to do the same thing, it would cost too much money. So you have these nice simple systems, right? Economics always plays a role in process control. You want to make sure the application is worthwhile. Okay, so this type of closed-loop control uses what’s known as a PID equation. So I’m going to go ahead and get rid of this mechanical part over here. And we’re going to introduce the PID equation.

The PID equation is going to be covered in further videos, but I want to introduce it now. Because it’s what drives this automatic control down here, your PID equation is your controller output, CO. The new movement to the control valve is going to be equal to some bias value, which is kind of a basis. It’s where it starts from. It’s going to add to that a correction based on the error deviation between setpoint and measurement. So you’re gonna have your controller gain, times the current error of your system, plus the integral term. So this is proportional. You’re going to introduce your integral term, which is going to have these little tuning parameters in front of it as well, which we’re going to go over later times the integral of air dt plus, now your derivative term, which is again going to have this KC in front of it, tau d, which is your derivative, right? So this is your integral. This is your derivative times d of t d t. This is the PID equation. This is used extensively throughout the process industry. It’s able to take measurements, perform this map and compute where the control valve should be. This is automatic control. These tuning parameters are what we’re going to focus on in later video series. Thank you for your time today. This was manual versus automatic control.

If you have a particular topic or an idea that you would like us to cover, please email us at Thank you, and I hope you enjoy this video series.