The digital twin is a digital model that makes it possible to predict malfunctions and speed up the company’s experimental processes. Let’s see what it is, how it works and where it is used.
A digital twin is a virtual alter-ego of a physical resource with which it can exchange data synchronously or asynchronously. A digital twin can replicate many things: objects, processes, infrastructures and many more.
These digital models are highly appreciated for their predictive capacity and experimentation. By testing a complex system, such as a machine for making boxes, you can test its operation and predict any malfunctions thanks to artificial intelligence. In this way, the company that uses this technology can save a lot of money, because testing an object in virtual reality guarantees its integrity in the experimental phase, without the risk of damaging it. Furthermore, thanks to AI, it is possible to predict any unforeseen events due to operation.
Imagine you want to build a car: crash testing can be extremely expensive, especially since the car would then be destroyed. Simulating the same incident in a virtual space with a digital twin is much less expensive and avoids waste. This logic can potentially be applied to an infinite number of complex systems: buildings, processes in work machines, etc.
How a digital twin works
Ideally a digital twin contains all the information of the physical twin through a 3D representation of its aspects at all levels (electronic, mechanical, geometric, etc.).
Thanks to this information it is possible to develop a predictive model and an experimental. With the predictive model it is possible to simulate the behavior of a complex system which highlights potential errors and malfunctions; while with the experimental one, by testing a product, it is possible to save on the creation of an expensive physical prototype.
Digital twins can be of four types:
- Component/Part Twins: Component twins are the fundamental unit of a digital twin. These are parts that are essential to the functioning of an asset, such as the engine inside an automobile.
- Asset twins: when two or more components work at the same time, they form what is called asset. Asset twins allow you to analyze the interaction between physical parts, generating a lot of performance data that can be processed and transformed into insights.
- System Twins: The next level is system twins (or units), which help you understand how different assets combine to form a whole functioning system. System twins provide an overview of how assets interact and can suggest possible performance improvements.
- Process Twins: Process twins show how systems work together to form an entire manufacturing plant. Process twins make it possible to determine precise timing capable of influencing overall effectiveness to the maximum.
How to create a digital twin
The digital twin are made using sophisticated software, designed by leading tech companies such as Amazon AWS, Dassault Systèmes, IBM, Oracle, PTC and Siemens. According to McKinsey creating a digital twin is a complex process that can be divided into three phases: creating a project, building the initial digital twin and then increasing its capabilities.
In the project creation phase it is important to focus on your goals: what do you want to do with the digital twin? What data is needed to develop it in the best possible way? Where can you find the data you need? Once all the initial information has been collected, the engineers assemble the fundamental data, thus creating a first primordial model that allows the visualization of some use cases. Thanks to the data that is collected from the first experiments, it is subsequently possible to gradually start improving the digital twin.
Once the initial digital twin use cases are up and running, it’s time to expand its capabilities by adding more data and analytics to support new use cases. In this phase, companies use their digital twins together with artificial intelligence to simulate scenarios or possible situations.
Where a digital twin is used
As mentioned, the digital twin can be used in many different situations and is able to provide valuable insights to the companies that use it. General Electric, for example, has equipped its turbines with some sensors that communicate with their digital twin. In this way, the company is able to provide a very cutting-edge after-sales service, since the digital twin communicates with the physical one almost in real time, providing all the data needed to provide assistance if needed.
But that’s not all, the digital twin can in fact be used in healthcare, for example when a doctor has to operate. By making a digital twin of the patient, it is possible to experiment with procedures in a simulated environment, thus not putting his life at risk.
It can also be used to check the strength of a building before it is actually built. In fact, by creating a digital twin of a building (a building, a tower, a house) it is possible to subject it to shocks and bumps of various kinds, with the aim of verifying its resistance.
Finally, another very fitting example can be that of Formula 1, in which digital twins are used to simulate racing and try to predict car performance.
In Italy, among the companies that use the digital twin is Dallara, an important car manufacturer that uses this technology to test the performance of the cars it designs. In this way Dallara is free to change a component of a machine simply and quickly and test its performance in a safe environment.
Many companies, especially those specialized in industrial production, already use the digital twin, as they benefit greatly from its characteristics. In the future, it can be expected that this technology will not only improve but be used in more and more sectors.
Original article published on Money.it Italy 2023-01-21 15:20:00. Original title: Digital twin: cos’è, come funziona, dove si usa
