By Subhankar Pal, research and innovation executive, Altran
2016 was pretty momentous. Donald Trump won the U.S. presidential election and the UK voted to leave the European Union. Spotlight won best film at the Academy Awards and Stranger Things debuted on Netflix. There was nothing strange about a concept called digital twins that debuted on Gartner’s 2016 Hype Cycle for the internet of things (IoT). By 2018, it was up at the top of the hype cycle.
Having established itself within the aerospace industry and even helping NASA maintain its equipment up in space, the digital twins concept has quietly started to revolutionize wind operations, and it’s not showing any signs of slowing down. It’s given some of the energy industry’s biggest players the kind of accessibility, visibility and foresight that, until fairly recently, probably seemed like science fiction. This article looks at digital twin technology – what it is and how it’s being used to help harness the power of wind with much greater efficacy for a greener world.
Think of a digital twin as a digital “mirror” of a physical asset that can allow users to see what’s going on under the hood without needing to be physically present. The physical asset, perhaps an engine or a turbine, is built with carefully placed sensors that can collect real-time data and operational status. This information can then be sent via the cloud to a piece of software for interpretation and scrutiny, either in the form of a 3D digital representation or raw data. Machine learning (ML) and artificial intelligence (AI) can also be applied to recognize patterns, predict faults and highlight important variables that the human eye would otherwise miss.
Leveraging the latest technologies
In a world striving for cleaner energy, digital twins with AI and ML can automate the production, installation and maintenance of wind turbines for efficient power generation. It can do so using data from existing systems cost-effectively for cheaper power generation. Put simply, it can make engineers’ lives much, much easier. Instead of dispatching a team of human workers to diagnose a fault, engineers can analyze and identify problems remotely, only dispatching teams where necessary.
Using digital twins with augmented reality (AR), virtual reality (VR) and mixed reality (MR) can bring projects to life for remote engineers. They can fully interact, see each other in different locations, draw, highlight equipment associated with a fault and visualize how machines can operate together.
ML can recognize patterns and even self-diagnose when there are complications – alerting key personnel only when their attention is needed. This can strengthen what is called intelligence amplification. This refers to the use of technology to speed up processes, allowing engineers to accelerate decision making with fewer resources. This is where digital twins can really excel – especially when wind farms are in remote locations.
Before digital twin modelling, any machine fault would need to be diagnosed and fixed by whichever team happened to be in the vicinity, even if that team was not the best skilled. Now, if there’s a fault with, say, a device on an offshore wind farm in the Mediterranean and your expert is in California, that’s not a problem.
They can simply pull up a real-time 3D model of the device and access all the data they need to rectify it.
Benefits for safety, training and R&D
Then of course, there’s the health and safety element. Being able to see how a machine is operating without the need for “boots on the ground” has obvious benefits when it comes to keeping employees safe, out of harm’s way and away from unnecessary risk in hazardous environments.
Digital twins can also provide comprehensive training for new engineers. For example, it can demonstrate how to carry out standard operating procedures like removing a part of a turbine blade and installing the replacements accurately. Multiple trainees can easily participate in the session collaboratively and even use interactive displays such as AR or VR. What’s more, it can also significantly lower the skill bar, enabling people with only basic knowledge to make decisions based on top-level feedback.
Aside from the myriad benefits of monitoring machinery, digital twins can also have profound advantages when it comes to R&D. All the data that’s gleaned from existing equipment can be used to model future designs — improving processes, highlighting weaknesses and fixing inefficiencies along the way. This ability to accurately model a machine before it’s manufactured has huge cost-saving potential, and the more a business uses digital twin technology, the more data it will have to play with.
Needless to say, this all requires a lot of data.
Until recently digital twin technology was only available to businesses that had the ability to collect, store and process substantial volumes of data. This, at least in part, is one of the reasons digital twins remained more or less in the shadows after the technology was conceived in the early ’00s. Businesses simply did not have the big data clout needed to take full advantage. With advances in computing power, that changed. Thanks to faster cloud computing, the internet of things (IoT) and the affordability of machine learning algorithms, digital twin technology is more accessible than ever.
Implications for wind farms
For the wind industry, the true value of digital twins lies in its ability to monitor the condition of an entire fleet of turbines, regardless of geography. This has innumerable benefits, but in particular it allows wind farm operators to proactively plan maintenance visits, reduce labor costs, limit down time and identify inefficiencies. In other words, it gives engineers 360° visibility and foresight that, until very recently, was impossible.
A digital twin can also give wind operators a unique perspective on individual components, tracking them through the entire unit’s lifecycle. For example, if a fault is already being tended to in the field, a decision might be made to swap one or two other degrading components at the same time. This reduces the number of engineer visits and makes the monitor and repair process efficient.
Modelling is also key here. Data modelling can aid in the manufacturing process, but it’s also invaluable when it comes to playing out “what if” scenarios. For instance, wind farm operators could use a digital twin to accurately predict what might happen if they increased the power rate of their turbines. Would the electricity generated at a particular rate be worth the strain on the components? The ability to rapidly play out these cost-benefit models can optimize the full fleet of turbines.
Having made its mark with the likes of NASA, digital twin technology is enjoying widespread use in a number of industries — from oil and gas and automotive to life sciences and urban planning. According to Gartner, 75% of organizations implementing IoT already use digital twins or plan to within a year. Within the wind energy industry, the potential is huge. Of course, it’s one thing to adopt the technology, it’s another to deploy and integrate it effectively. By understanding how other sectors have used and benefitted from it, wind farm operators can harness its potential as the cornerstone of a green energy revolution.
Subhankar Pal is an Assistant Vice President of Technology and Innovation at Altran, the global leader in engineering and R&D services. Altran was acquired by Capgemini in April 2020.