Hybrid Twin™ vs. Digital Twin: We'll Tell You the Difference and Which Can Save the Life of Your Asset
Digital Twin and Hybrid Twin™. You may have heard these seemingly synonymous terms before, but do you know the difference between the two? In an interview with Dr. Slim Soua of TWI, an independent research and technology organization, he shares with us the subtle yet important variances in them as well as how using one of them can save your company tens or even hundreds of thousands of dollars.Thursday, September 5, 2019
By Rajab Said
ESI and TWI are expanding their collaboration program. Can you tell us more about the expansion and what it means for companies worldwide trying to perserve the life of their assests?
ESI and TWI have been collaborating for several years, focusing mainly on typical applications of Computer-Aided Engineering (CAE) in the domain of Materials and Welding Processes. The collaboration has included utilizing existing software capabilities for our in-house research: such as investigating challenging metallurgical & mechanical issues encountered during welding with ESI SYSWELD; or leveraging the crash & strength models in ESI Virtual Performance Solution to develop new concepts for increasing the safety of structures fabricated in composites. We have also worked together on developing new and innovative solutions through large scale collaborative R&D projects, for instance the EU project SIMUTOOL which required a simulation platform for manufacturing Composites via Microwave Heating.
Over the last 40 years, TWI has grown significantly and diversified its expertise well beyond materials and joining technologies, including the development of world-leading capabilities in structural integrity management. The Integrity Management Group (IMG) now offers state-of-the-art monitoring solutions, which is helping hundreds of companies worldwide manage the life of their asset, avoid engineering failure, improve safety and reliability, reduce inspection & maintenance costs, ensure regulatory compliance, and optimize operating expenditure.
Thanks to the ongoing collaboration, we at the IMG team became aware of ESI’s strategic plan to expand their solutions beyond Virtual Prototyping to cover the entire Product Performance Lifecycle™ (PPL). To us, a move like this means that ESI is no longer only providing solutions to address issues for the design and manufacturing communities but will now need to assist customers with the challenges associated with the product “In Operation” as well. This creates a clear synergy between ESI’s vision and TWI’s capabilities. Additionally, the role the Industrial Internet of Things (IIoT) will play in shaping the future of business is overwhelming. The Industry 4.0 initiative, which started in Germany in 2012, is now a global phenomenon and ESI and TWI are set to be a part of it.
The British government, like others in the developed world, is investing in large scale R&D funding programs to support new developments based on the IIoT. In response to a recent call from Innovate UK (Emerging and Enabling Technologies Round-1), ESI and TWI have worked together, with two local SMEs (Agility3 & Dashboard) and a specialized research group at the Brunel University London, to submit a proposal (“WindTwin”) for the development of a comprehensive Hybrid Twin™ solution to be used across the Wind Energy sector. Our bid was successful, the Project started recently and is scheduled for completion in December of 2019.
You mention the “Hybrid Twin™”. How is that different from a “Digital Twin”?
This is a great question as the field is still in its inception and the terminology can be confusing.
The term Digital Twin is being used to refer to the digital replica of physical assets, derived from real-life data collected using various types of sensors and monitoring technologies of the asset while in operation. It often means that an analytical data-driven model (i.e. the twin) is built to analyse, update, and/or manage the performance of its physical counterpart. It can use a range of tools and advanced algorithms - - such as Machine Learning (ML), decision making, or even Artificial Intelligence (AI) technologies - - to analyze and visualize the “Big Data” collected.
In contrast, the Hybrid Twin™ is associated with solutions where an additional, complementary virtual model is built. This supplementary model is necessarily physics-based and describes cause and effect relationships.
By its own nature, a Digital Twin - based solution is limited by the number and location of available sensors, as well as by the quality of data collected. For example, we know from our preparations for the WindTwin Project that a typical wind turbine may include six to eight strain-sensors per blade, and one or two temperature-sensors in the gearbox, etc. Such sensors provide a 24/7 stream of data, which is indeed very valuable for monitoring purposes and can be utilized to build, validate, and improve data-derived models. On the other hand, wind farm operators are keen to learn more than just the mechanical and thermal behavior at the locations of these sensors. They also would like to be able to anticipate the full range of potential consequences when any of these sensors starts sending an uncommon pattern of data, and to have a tool that will enable them to investigate potential actions to avoid failure or degraded performance. Ideally, they would like such a tool to recommend the best intervention; one that minimizes the interruption to operation and also the total cost. Here, the role of high-fidelity
physics-based models is crucial and the advantages they bring to the Hybrid Twin™ are very clear.
How do you see the impact of the Hybrid Twin™ in relation to the WindTwin project, and beyond?
The WindTwin project aims to streamline the monitoring and maintenance processes for wind farm operators. Addressing the needs of both onshore and offshore farms, the goal is to increase the availability and reliability of wind turbines. A dedicated Hybrid Twin™ based solution will incorporate all relevant physics at the sub-systems level, supported by sufficient degradation models. Through the solution, operators will be able to use the WindTwin to diagnose performance variations, and deploy condition-based maintenance, instead of pre-determined, schedule-based strategies. This will undoubtedly reduce downtime, inspection & maintenance costs, enabling operators to virtually test maintenance upgrades before deployment. It will also help better control wind turbine settings in order to optimise performance and energy output.
Worldwide, the total number of wind turbines reported by the end of 2016 was just below 250,000. The market size for operation and maintenance is estimated at about 10 billion USD today and is expected to double by 2022. Under normal operation, the annual maintenance cost is about 5% of the capital investment, but it can be as high as 10% in some cases. The potential economic impact of effective use of a Hybrid Twin™ is clear.
Of course, the increasing pressure to eliminate failure, reduce maintenance cost, optimise performance, etc. is not unique to wind turbines. The list of potential applications goes across almost all industry sectors, such as Heavy Machinery, Aerospace, Defense and Marine.
Learn more about Delivering the Hybrid Twin™.
TWI is one of the world’s foremost independent research and technology organizations, with expertise in materials joining and engineering processes as applied in industry. TWI specializes in innovation, knowledge transfer and in solving problems across all aspects of manufacturing, fabrication and wholelife integrity management. Established in Cambridge, UK in 1946, the organization has gained a first-class reputation for service through its teams of respected consultants, scientists, engineers and support staff. For more information www.twi-global.com
Academic and R&D Collaboration Manager
Rajab has more than 25 years of experience in computational modeling, covering both fundamental & industrial research as well as technical project management and business development. He spent 12 years at the Civil & Computational Engineering Centre at Swansea University, studying for an MSc and a PhD, working as a senior research assistant for three years and as the technical manager of the Centre of Excellence for four.
In 2008, after two years as Business Development Manager at Simpleware (now part of Synopsis), Rajab joined ESI Group as their technical business development manager for ESI UK. In 2014, he took on the role as academic and R&D collaboration manager for ESI Group worldwide. During his 10 years at ESI, Rajab managed several collaborative R&D projects - addressing emerging challenges within the Virtual Manufacturing domain (including Casting, Sheet Metal Forming, and Composites) – and he is currently leading an R&D consortium working on the development of Digital/Hybrid Twin solution for the Wind Energy sector.