Before we dive into virtual sensors, we need to cover some foundational concepts around the Digital Twin. With this technology, you have sensory data that you’re capturing from a product, and you’re feeding it into a simulation. The simulation will be running very fast, and will often be a systems simulation; a 1D simulation that will have just equations as part of a diagram. It may be a reduced-order model, or it may be a multi-body dynamics model.
What Does It Do?
By definition, a virtual sensor is a type of software that, given the available information, processes what a physical sensor otherwise would. It learns to interpret the relationships between the different variables, and observes readings from the different instruments. Think of it as a kind of a “ghost” of the physical sensor.
The idea here is that if you’re running a simulation and you’re getting accurate inputs, and it is mimicking the behavior of a product in real life, in real time, you can then instrument that simulation model with taking measurements at different locations. You are going to be getting readings from that virtual sensor, which is where you’re taking measurements, and then you can actually use that to complement physical sensors.
Why is it Important?
What’s great about this is you can place virtual sensors on a simulation model anywhere. That’s a lot different from what you see in the real, physical world. In many cases, you might not be able to place the physical sensor exactly where you want it to be.
For example, think about locations with high radiation. You’re not going to be able to get close enough to place a physical sensor. It will be fried very, very quickly. Or maybe it’s a mechanical assembly, and there isn’t enough clearance where you want to place the sensor. Other factors that influence the decision to choose virtual over physical include cost and reliability.
The Digital Twin can help deliver that, and you don’t necessarily have to be connected to take advantage of virtual sensors.