If you’ve been following our articles, you probably noticed us talking a lot about control systems, a central topic of industrial automation. We talked about control theory, how controllers work, and now it’s time to talk about control strategies.
With the knowledge we shared here, you can probably design a simple closed-loop control system. But the processes in industrial automation are usually not as simple as making toast or cranking the AC up.
Sometimes you have to control multiple variables to keep your primary process value (PV) within the desired range. You may also come across disturbances in your plant which can affect your readings and thus your entire process.
For these more complex situations, you have a few control strategies that can simplify your process or ensure more control in your plant. In this article, we’ll check out some of these strategies, which we also call advanced regulatory control (ARC).
Let’s start with feedforward control. If you have a measureable disturbance in your system, then this strategy can reduce the disturbance’s effects to improve your system’s performance and speed.
In feedback-only control, the controller will only act when it detects an error after your PV changes. With a feedforward control, you measure the disturbances and can act before errors affect your PV.
Let’s have an example to understand it better. Imagine you want to control a heating exchange process with cold water and hot steam in the exchange. In this system, you control the water temperature by changing the steam flow in the heat exchanger.
If the intake of cold water increases, then the temperature at the outlet should decrease. However, with a feedback-only controller, the control system won’t act until it sees an error. That means that the temperature at the outlet will have already changed.
With a feedforward controller added to the system, you can measure the water intake as a disturbance in the system. By doing so, you can predict the temperature change in the outlet before it happens. This way, your error will depend not only on the water outlet temperature but also on the water inlet flow rate.
Ratio control is a specific feedforward-control strategy. You’ll mostly see it used to control the ratio of flow for two streams.
Consider the control of the fuel-air ratio in burners. Here you’ll measure the flow of fuel and air and feed the ratio of the two to your controller. This controller acts in just one valve and has your desired ratio as its SP.
Now you can consider the flow of one fluid your measured value and the other your disturbance. Your controller measures your disturbance, calculates the ratio, and adjusts the valve if necessary to increase or decrease the ratio of the fluids.
You can also use cascade control to deal with disturbances. Just like the feedforward control, the cascade works predictively -with one small difference.
On a feedforward control, you measure the disturbance and feed it to the same controller that controls your primary variable. In a cascade controller, you have two controllers, with the output of one as the input of the next. Before you get confused, let’s whip out an example.
Let’s stick to the heater exchange, but this time, we’ll use the steam flow rate as our disturbance. When the steam pressure changes, the flow can change even if the control valve hasn’t moved. To avoid this disturbance, you can add a flow controller (FC) in series with the temperature controller (TC) to complete the control loop. This setup makes the output of the TC the input of the FC. We call the flow control loop the inner loop and the temperature control loop the outer loop.
The usage of a cascade control strategy brings two main benefits to your application. First, it can eliminate the effects a disturbance can have on your system. Second, it improves the overall dynamic performance of the entire loop.
Split-range control is the go-to strategy when you have multiple variables (inputs) to control a single output.
Let’s say you have a pH control system that must maintain a solution at pH 7. Using a split-range control strategy, you can have the controller operate two valves, one for a basic solution and one for an acid solution.
If you have pH 7 as 50 percent of your total range, when the measured value falls below that range, then the controller will open the valve to the basic solution. If your measured value goes over 50, then the acid valve will open. And if the pH reaches 7, then both valves will close and remain so until the value changes again.
Selective or override control
Last but not least, we have the override or selective control strategy. You’ll usually find override control when you have one controlled value but multiple measured values. That requires more than one controller in the loop, but only one will actively control the CV.
For this setup, you’ll need selector switches, which come in two types, the high selector switch (HSS) and low selector switch (LSS). Which you use will depend on your application.
Going again to the boiler, imagine now that we need to control the pressure in the discharge line. You can control that pressure by controlling the valve in the discharge line. However, you must also keep the water level inside the boiler high enough to immerse the heating coil.
Here you’ll use an override control strategy with a low selector switch. A level transmitter will monitor the water level inside the boiler, and a pressure transmitter will monitor the pressure in the discharge line.
If the water in the boiler rises above the minimum value, then the pressure in the discharge line will control the valve. If the water falls below the lower limit, then the LSS switch will close the valve in the discharge line, regardless of the pressure.
Feedback control systems and PID controllers work well to control simple applications. However, more complex systems may have more disturbances, which can decrease the performance and speed of your process.
If you know and can measure the disturbances in your systems, then you can use ARC strategies to improve your process’s overall performance and reliability.
This article reviewed the most common advanced control strategies out there. Nonetheless, you’ll find more advanced strategies to improve your application. Many manufacturers offer the design and implementation of these strategies as a service, if you want to look into them.
To know more about control strategies, you can get in touch with our engineers and we will be happy to help.