Digital Twin Technology
The process industries are embracing advanced digital technologies in order to optimize assets and processes for greater agility, resilience and sustainability. The digital transformation journey is well underway for most manufacturers. For many, digital twin technology—creating virtualized copies or digital representations of physical assets and operating behaviors—will play a key role in this journey.
A digital twin isn’t just a model; it can be manipulated and adjusted like a real-world object, and it will behave like the modeled asset. Digital twin technology (also known as process simulation technology) seeks to reproduce an object’s smallest details, and digital twin software allows the various components of an object to interact realistically.
An ideal digital twin would provide a perfect simulation of an asset from the smallest to largest scales. With current technology, this is limited by the amount of computing power that can be applied to the problem, as well as the limits of our understanding of how materials function. Nevertheless, using digital twin technology allows for significant advancements and improvements to industrial processes.
Digital twin technology allows process changes to be modeled and explored in digital twin software, without the disruption in production that can result from adjusting actual asset operating parameters. This provides enterprises with an essential tool in production optimization, plant digitalization, and industrial digitalization.
From the earliest days of digital computing, companies have used software to simulate physical processes. Even when enterprise computing involved large and expensive equipment, it was often more efficient to use process simulation software that could model potential modifications to production rather than risk critical infrastructure on experimentation.
Digital twin technology is based on the initial idea of engineers and programmers to use information about the design and operation of the asset or process being modelled. The twin is kept up to date with operating data supplied by technicians and sensors monitoring the digitally twinned process or asset.
The creation of these process simulations requires the expertise of plant engineers and operators, as well as the computer programmers that actually encode the software. The plant engineers and operators provide the technical data about the real-world functioning of a given asset, and the programmers find ways of representing that data dynamically in software.
With the significant improvements in computing price-performance, process simulator software has increased in value as it has become faster and more powerful. The explosion in sensing capabilities through cheap, networked industrial sensors has made it possible to have a process simulation that incorporates live data from the process that is being modeled. What would previously have required laborious data-entry of production records can now be automated and updated in real-time.
Digital twin software can both model hypothetical changes and actual past performances. For hypothetical changes, the software uses the relationships between components to model future situations; for observing past performances, it searches through the database of historical data to find situations in which the changes have already occurred.
For example, if plant operators want to know what will happen when a compressor is operated at half pressure, the digital twin software can either make a prediction, or it can search for times when the compressor operated at half pressure and simply retrieve the data generated under those conditions. If the exact conditions exist in the historical data, this information may be more accurate than a simulated prediction. However, if there are multiple variables being adjusted, it may be more valuable to find a predicted outcome.
Companies use digital twin technology to accurately model industrial equipment and processes. The advantages of creating a simulation of an asset in digital twin software are numerous, and often far outweigh the initial cost of creating the model.
A digital twin allows for close monitoring of an asset or process, revealing ways in which an asset may be used more efficiently. Inefficiencies that might be difficult to identify can be made clearer in a digital twin. For production processes that involve large footprints, digital twin software may provide a concise overview that would otherwise be unavailable. For example, it would be impossible for one person to appraise the activity across the entire line of a refinery that covers multiple city blocks.
In addition to increased observation, digital twin technology gives plant operators and process engineers the chance to run experiments on accurate models and predict how various changes will affect production. A digital twin has the distinct advantage that, should the requested operating parameters lead to asset failure in the simulation, the digital model can simply be returned to an earlier state with a click of a button. This lets operators tweak and adjust the conditions of the digital twin without risking the loss of a physical asset or causing costly production downtime.
Why would a company use digital twin technology?
Any company that operates industrial processes that involve valuable assets and equipment has to contend with the problem of how to increase efficiency while keeping production online. Adjusting the operating conditions of an asset risks disrupting production, and this can lead to overreliance on past parameters for current operations. There is no guarantee that the current working solution is the most efficient solution; it could just be the set of parameters that turned a profit.
Without a way to simulate the assets being optimized, the experimentation required for production optimization is significantly hindered. Using digital twins to model and predict how an asset will respond to various conditions can increase efficiency as well as help plan for extraordinary situations.
What does a company need to implement a digital twin?
A digital twin needs to be created with knowledge about the fundamental physical properties of the asset being modeled. Many software suites include extensive materials lists that allow for the rapid creation of digital twins. Digital twin technology uses this information when simulating production processes, both to contain the simulation in the actual operation envelope, and to help model deviations from the norm.
An extensive suite of industrial sensors to collect data from the asset being digitally twinned is required to ensure that the digital twin provides an accurate reflection of the asset. This data can either be stored and input to the model afterwards or networked to stream the information to the digital twin in real time.
The Digital Twin and the Smart Enterprise
Hybrid Modeling AI and Domain Expertise Combine to Optimize Assets