What Is a Digital Twin?
August 10, 2017
GE Reports Canada
Digital Twins are changing the way we do business. Here’s what you need to know about this critical trend in industrial tech.
What is a Digital Twin?
Digital Twins are virtual counterparts to real-world assets. Using sensors and other sources of information, engineers create computerized duplicates of equipment, machines, or facilities. Digital Twins can also be made for physical sites, before an asset has even been built.
What are Digital Twins made of?
Digital Twins synthesize data from several key sources: manufacturing information, physical information, operational information gathered from sensors, insights generated by analytics software, and artificial intelligence algorithms that continually improve on those learnings. The Digital Twin brings together everything that can be known about the asset, integrates all that intelligence into a physics-based model, which engineers then use to learn things they otherwise couldn’t have discovered.
Does a Digital Twin look like its real-world counterpart?
Sometimes, but not always. Operators may use Digital Twins to check the form and fit of an asset, in which case it will include a visual representation. But if the Digital Twin was developed for operational, process, and performance representations, then there’s no need for it to generate a copy of the physical design.
How does a Digital Twin work?
Think of a GE wind turbine. There’s a real one that sits out in the field, capturing wind energy. But there’s also a digital model of that specific turbine, which resides in the Cloud. The digital model includes material information like the dimensions, manufacturing details, etc, as well as key environmental information, like geo-location and elevation.
Those two entities—the physical wind turbine and the virtual one—become a Digital Twin when a network of communication is opened between them. Each wind turbine is equipped with over 150 sensors that monitor things like thermal dynamics, mechanical functioning, electrical states, and more, not to mention environmental conditions like temperature, humidity, wind, etc. By transmitting the data from these sensors to the digital model in the Cloud at ultra-high connection speeds, a Digital Twin is created.
How is a Digital Twin used?
There are many ways that Digital Twins can be useful. For example, since the Digital Twin incorporates sensor data from the operation of the asset, engineers can use it to directly monitor the life of the physical asset and predict when maintenance will be required. One way they do this is by comparing the Digital Twin to an optimal performance model to see if there’s any discrepancy.
Engineers can also run simulations on the Digital Twin to see if there are more efficient ways to operate the asset. Not only that, they can use Digital Twins to design new products in the virtual environment.
Think of it like having an indestructible stunt double. The operator can use the computerized counterpart to test the asset in different ways and investigate new operating strategies. If the simulations don’t work on the Digital Twin, there’s no harm done, but if they do, then the operator can implement the improvements and outcomes will be that much better.
Digital Twins aren’t just for machines. It’s also possible to create precise digital replicas of real-world locations that enable companies to reduce cost-overruns and delays from capital projects. Digital Twins can be created for entire facilities to make operations more efficient.
Many of these applications for Digital Twins are available on GE’s Predix platform—and many more applications have yet to be created. Predix offers software tools for companies to customize Digital Twin technology to their specific needs and amplify the benefits to the greatest extent.
What happens when you have multiple Digital Twins?
Digital Twins can also be linked up as a network. At its most basic level, a Digital Twin is a one-to-one line of communication between a real-world machine and its virtual counterpart. Think of how much operators can learn by creating a Digital Twin of a single wind turbine. Now imagine a fleet of thousands of wind turbines and their Digital Twins coming together in a larger network. That’s a lot of data.
The artificial intelligence algorithms used in Digital Twins are hungry for just this sort of massive data set. Using advanced analytics, it’s possible to compare all those turbines and produce insights to get the most out of every turbine.
Even better, the network of Digital Twins of wind turbines can be connected to other networks of digital twins for different types of mechanical equipment, such as the grid that transports the energy to consumers or storage systems that house the energy.
What does a Digital Twin do?
Digital Twin technology sees: It gathers data about every aspect of the asset from many different sensors.
Digital Twin technology thinks: It run thousands of simulations using data from across the fleet and throughout the larger network, along with masses of historical data, and makes informed suggestions with predictable levels of confidence in the outcome.
Digital Twin technology acts: By relaying critical information to operators or by automatically executing what needs to be done.
What are the benefits of a Digital Twin?
Using algorithms that compare and simulate, it becomes possible to boost efficiency and reduce downtime. By comparing all the Digital Twins in a network, it becomes possible to optimize the operations of an entire factory. And once the functional capabilities of a factory are improved, it becomes possible to implement even better business capabilities. That is, digital twins not only optimize machines—they also optimize business models.
There are as many benefits of Digital Twins as there are industries where Digital Twins can be generated. In healthcare, Digital Twins of the human body are starting to be made, to improve patient monitoring. Digital Twins have been used to create megawatt-sized circuit breakers to guard against prolonged outages at power companies. In the in situ oil sands sector, Digital Twins can be made for active wells using GE’s Adaptix software, allowing operators to minimize their use of both natural gas and steam. These are just a few examples.
What’s an example of a Digital Twin?
GE makes Digital Twins of jet engines. By analyzing a constant stream of information from sensors on jet engines and comparing that to fleet-wide historical data, complex algorithms can calculate the best time to do maintenance. With that information, airlines can make better decisions about how their planes are used. GE Aviation is creating smart jet engines that can log into the web and report to a ground-based maintenance application while in flight.