Process Control Systems
Note
Process control is important in many processes to ensure efficient and safe operations.
Feedback Loops
A feedback loop is one of the most basic forms of process control. It has four basic components: a sensor, a controller, an actuator, and the process itself. The sensor monitors some kind of important information. The controller is the logic of the system. The actuator is the physical method of manipulating the process. The process is what is trying to be controlled.
For example, consider controlling room temperature. This is a feedback loop. A thermostat acts as a sensor for the room’s temperature. The software is the controller. The heating vent in the room is the physical means to manipulating the process. The room temperature is the process.
In a feedback look, the controller logic must run continuously. This is because the process is always changing.
The frequency, or how often the loop runs, depends on the desired level of control and the sensitivity of the system. For example, room temperature changes slower compared to other processes, so it is not necessary or useful to have a high frequency, such as every microsecond.
Feedback loops can often be improved. Consider the example above of room temperature. Let us improve that loop by introducing a proportional controller. A proportional controller calculates an error value as the difference between the desired setpoint and a measured process variable and then applies a correction based on proportion. In this case, a heating vent whose position can be controlled could read an error, and open the vent according to how big the error signal is. As the temperature comes closer to the setpoint, the heating will slow down so that the setpoint can be approached gradually. (Proportional controllers can also have undesirable effects, but this runs outside the scope of this course.)
Other Variations
Other variations exist on process control.
A feedback loop is an example of a closed loop, as information from the process is used to control the same process. An open loop is one where the process is controlled without monitoring the process. Open loops cannot deal with changes in the process or check itself to see if it is succeeding.
Feedforward control occurs in systems with processes in series. Information from an upstream process can be used to control a downstream process. Feedforward control requires a good model of how a process responds and deals with unknown events. These systems are often used in conjunction with feedback loops.
For example, a flood protection system could use feedforward control. If an upstream monitoring station detects large flow rates, then the controller for the flood protection system could be signaled. The controller asks the actuator, a gate, to open and divert water into a reservoir to keep water levels manageable.
More complex problems will have many sensors and ways to control the process. Process control architecture for complex problems will therefore be more complex. Below is a diagram for a complex example of a self-driving car.
This architecture is based on a MAPE-K structure, which has four major steps: monitor, analyze, plan, and execute. All four of these steps must have knowledge of the process.
| Step | Description |
|---|---|
| Monitor | The component or collection of components that interfaces with sensors, turning raw signals into meaningful information for the software. |
| Analyze | The software must next analyze this data. Data could be integrated from several sensors, so analysis might be quite sophisticated. More sophistication generally requires a better model of the process. |
| Plan | Next, planning needs to be done. More advanced planning needs will require more knowledge. This knowledge is in the form of a model of the system. Such models should be designed in close conjunction with engineers specializing in the process. |
| Execute | The final step is to execute. Analysis and planning should make execution relatively simple. They will simply turn desired actuator positions into signals to communicate to those actuators. |
Knowledge is very important in MAPE-K systems, particularly complex ones. Currently, technology companies are investing heavily in such systems, due to the advent of machine learning and big data. This investment goes towards real-world testing, machine learning, and data modelling in order to operate these systems. Effective software architecture is key. Even though it is handled by specialized professionals, it is important to be aware of process control. Software is becoming a bigger part of controlling physical systems.
