Updated: May 21, 2020
First off, I don’t think anybody wants to clone me (and if… double the trouble I’d say). But I understand the benefits of creating a digital copy of a physical system, a so-called digital twin. Let me give you a further definition: digital twin is a buzzword in the industrial automation world and is best described as the flawless integration of data between a virtual and a physical system in either direction.
The idea behind this is actually very clever: in virtually testing certain components of the physical system on the digital twin, the factory floors can still operate, while the new control algorithms are being tested on the digital copy of that exact same factory floor. This is not only very efficient, but also less time-consuming and the company is not losing money. It goes without saying, that the digital twin must operate as close to reality as possible, so sophisticated scenarios can be realized. In continuously testing and verifying the algorithms, engineers can increase the quality and functionality of the simulation model. Furthermore, they can detect coding errors at a very early stage of the development process. And yes, what applies to a tractor (I have been talking about tractors and Hardware-in-the-Loop (HIL) simulation in my last post) also applies here: The earlier a coding error is detected, the better. There are further benefits to this, such as not having to worry about late failure detection or having to compensate for the lack of availability of a physical (real) test system.
If we stay with the concept of a digital twin a little longer, there is actually another case we can look at: a digital twin can also be made of a system (this could be a machine, a sensor, whatever the use-case may be) which does not exist in reality. This is the case when the digital twin is used as a design model for future products. You can probably tell by now, that digital twins can take any shape or form. Therefore, it can also be based on a behavioral model in the field of system development, a 3D model or a functional model depicting mechanical, electronical, and other properties and performance characteristics of the real twin (remember there may or may not be an existing twin) as realistically and comprehensively as possible. Which leads me to another very crucial term we need to look at: welcome to the world of model-based design!
But before I start writing about the wonders of model-based design, let me ramp up the digital twin concept for you. In my opinion, three main aspects are very important here: a digital twin is a copy of a real system or the further development of one, it exchanges data and digital twins bring massive benefits, if implemented correctly. Sounds awesome, right?
For even more facts about this, watch the following video and be sure to stay curious!