The digital twin family explained

28 November 2022

Naveen Kumar and Junichi Watanabe explain more about the different elements of digital twin technology and how it can offer benefits for the process sector.

A digital twin is a representation of a human, device, system, or process that replicates an actual process and has full knowledge of its historical performance. Digital twins drive agility and the convergence of understanding to enable effective decision-making and to determine strategies in order to maximise safety, reliability, and profitability by simulating devices, systems, or processes to forecast future performance.

A digital twin captures data to determine real-time performance. This data can be used across the entire life cycle of an asset for optimisation and predictive maintenance. 

Digital twins are able to drive improvements through advanced data analytics and operational insight; guide day-to-day decisions with in-depth, accurate data; utilise massive amounts of plant information to enable better decision-making; and ensure that actual performance meets planned performance.

The current digital twin solution for plant – which is a virtual replica of an actual plant –consists of a process digital twin and an asset digital twin. Both of these twins link with real plant operations and handle the data of the real plant and are continuously updated with real conditions. They are able to replicate reality, simulate it, and optimise the activities revolving around it. As such, digital twins constitute an evolving digital profile of the historical and current behaviour of a physical object or process, and this profile can help optimise business performance.

The process and asset digital twins will be closely connected in an integrated solution. Asset integrity information is part of the feedback loop to operators, giving full visibility of process impacts on mechanical assets, while process data is a critical input of predictive maintenance algorithms, allowing the recognition of process patterns which could lead to asset failure.

Asset digital twin
An asset digital twin is based on a 3D model as a repository of asset information and of cumulative data completed by properties and documents, such as manuals or operating procedures generated during all life cycle phases of the asset. The twin encompasses technical documents produced during Engineering Procurement and Construction (EPC). It expands with maintenance plans developed by vendors, and it is continuously updated with maintenance and inspection reports. The fundamental backbone of digital-enabled solutions during plant operations, it is supplied with key information during the EPC cycle at a marginal cost.

Benefits of an asset digital twin will include:

• Improved reliability and availability of assets by decreasing plant upsets due to human errors in planning and execution.
• Reduction in maintenance time, effort, and costs by increasing the reliability and productivity of day-to-day activities.
• Reduction in turnaround duration through better planning of activities on a 3D model, thus increasing plant availability.

Process digital twin
The process digital twin is becoming a key enabling factor for the new technologies connected to plant operation and optimisation, through the integration of the process model, the process optimisation engine, and real-time data from the plant. The process optimisation engine leverages thermodynamics to simulate the process and optimise its operations. As a further improvement, it is also possible to enhance and reinforce the thermodynamics model with machine learning algorithms to simulate the process and optimise its operations as precisely as possible, integrating production programs and economics in order to maximise the plant margin.

Process digital twin benefits include:

• An increase in plant margin and productivity by support operation in day-to-day activities, optimising the plant operating conditions, fast de-bottlenecking, and reducing the utilities consumption.

• Support for operation decision-making by the application of the operator training simulator.

Individual point solutions with digital twins do exist today, serving different purposes such as fit-for-purpose simulation models and individual data sources. But a future digital twin will be a one multi-purpose digital twin, which aligns the asset life cycle and value chain, a Multi-Purpose Dynamic Simulator (MPDS), and ubiquitous data sources. It is unrealistic to assume this future state can be achieved in one step, but it becomes more likely through the connectivity of valuable high-performing individual elements.

Just one example to demonstrate the benefits of a digital twin Application would be for a corrosion prediction and digital advisory system. 

If there is a gas leak in a facility the first question will always be ‘where is the leak?’ Then more questions arise – How can the leak be stopped? Why did it happen? And how can the plant be safely shutdown to avoid an accident?

Once the situation has been controlled controlling the situatin, a team will be tasked with identifying the reason for the leak and the to identify the losses plant incurred.

Corrosion is a slow poison to plant equipment and pipelines and its effects are often only apparent when the system has already sufficiently corroded – with destruction either imminent or already having occurred.

Yokogawa Electric and KBC have developed a hybrid solution that combines Artificial Intelligence (AI) based on Yokogawa Machine Learning techniques and KBC’s Petro-SIM simulation model with the OLI Alliance Engine to provide a corrosion condition prediction system that allows operators to monitor and manage systems, such as Health Index and corrosion rate.  

This system is also able to predict future corrosion issues  which enables operators to become proactive with corrosion management. 

Application of the AI/ML model  first requires the model to be put  online to take the current operating data at real time from plant through the distributed control system (DCS). Then, it  is then able to predict the future problem by showing the health Index (Positive = OK indication, Negative = Bad indication for future). It can also provide the top three ‘cause variables’ which are directly related to the cause of corrosion. Finally, it can be used to simulate the top three ‘cause variables’ to identify a mitigation plan to avoid future corrosion problems.

Naveen Kumar is a senior software engineer and Junichi Watanabe is an advanced solution general manager at Yokogawa Electric Company.


Contact Details and Archive...

Print this page | E-mail this page