Digital Twins: More than just visualisation
Digital Plant Specialists Sdn Bhd has developed a cloud or server based digital twin framework, incorporating our world leading ecosystem of technology partners, enabling companies to head towards operational excellence by helping them take a model-focused approach that quickly turns massive amounts of data into wisdom that generates business value. These powerful data insights mean asset failure can be predicted, hidden revenue opportunities can be uncovered, and businesses can continuously improve in the ever-changing, competitive market.
Digital Twins – Why are they necessary.
A digital twin is a representation of a physical asset that has a level of completeness and accuracy and includes context information that allows the user to understand its behavior and performance. DPS differentiate project digital twins used in engineering and construction, and performance digital twins used in operations and maintenance of existing and mature assets, both onshore and offshore.
The basis for a digital twin is accurate, up-to-date and accessible asset information. Properly managed and deployed, the digital twin changes over the asset lifecycle to reflect the asset changes such that the user can see crucial information about how the physical asset is or was performing in the real world. On top of this basic digital twin, more sophisticated aspects of the twin can be constructed, such as engineering calculations, predictive analytics and construction modeling for the modernisation and maintenance of the plant.
The digital twin will enable companies to head towards operational excellence by helping asset operators and owners take a model-focused approach that quickly turns massive amounts of data into wisdom that generates business value. These powerful data insights mean asset failure can be predicted improving the health and safety of your workforce, hidden revenue opportunities can be uncovered, and businesses can continuously improve in the ever-changing, competitive market.
The Digital Twin 3D Model
The different aspects of the digital twin are highly interdependent and their accuracy and currency impact on each other’s performance, and thereby project or plant performance. Therefore, the integrity and quality of data making up the digital twin must be managed efficiently through data governance.
To build a digital twin for an asset, an as built 3D model is created. This model can be based on current 3D models already in place with the operator or created through physical 3D laser scanning and modelling.
In many cases, the models that are available to our clients do not represent what is actually in the field, due to ongoing modernisation and upgrades to the plant. 3D scanning and model building will capture these changes and create an accurate representation of the asset, one that is required if you are to manage the plant efficiently.
Capture the Asset history
The history and wisdom of your operating asset resides in the legacy data, the operations, maintenance, construction, inspection, and engineering documentation generated over the life of the plant. Lacking a robust electronic data management system to maintain up-to-date critical information and operational data around assets during operations will incur significant waste on maintenance and operations budgets and resources to keep the assets running. Therefore, the key to achieving a digital twin is a comprehensive information management system that combines a centralised master data repository (MDR) with enterprise-wide information access and visualisation; one that allows users to have complete control of their asset information throughout the entire asset life and can be visualised with a 3D geospatial front end.
Through a process of intelligent data creation and validation, DPS utilises a suite of tools developed with the latest artificial intelligence and machine learning algorithms to extract data from key engineering data sources and formats including the engineering drawings, datasheets, lists, and records. This creates a very simple and structured data source which can be migrated into our EDMS platform through a tag system creating your master data repository (MDR). The validation process can remove document and drawing duplications, older revisions and non-essential data creating a single source of truth that represents your plant ready for linking with the 3D model.
Integrating critical data to the Digital Twin
Once the model is a true representation of the asset and the legacy data has been captured with the creation of the MDR, the digital twin can be linked to all the necessary attributes and engineering documentation – such as tag nomenclature, geometry, geospatial layout, design information of all components and process data now residing in the master data repository.
In addition, other business and safety-critical engineering data such as ERP maintenance and inspection data, procurement history, and real time equipment sensor data can be linked to the model.
Through a single data integrator, DPS can gather information and data around the asset, extracted from disparate data sources and validate them for accuracy against known standards to create viewable renditions of documents and drawings. This acts as a data validation layer to ensure that all data meet the correct standards throughout the asset lifecycle.
As the operational life continues, the digital copy is updated automatically, in real time, with current data, work records, and engineering information to optimise maintenance and operational activities. With this, engineers and operators can easily search the asset tags to access critical up-to-date engineering and work information and better understand the health of an asset.
Thinking Ahead – Predictive analytics
Previously, such tasks would take considerable time and effort, and would often lead to issues being missed, leading to failures or production outages. With the digital twin, operational and asset issues are flagged and addressed early-on, and the workflow becomes preventative, instead of reactive.
With the advent of advanced machine learning and AI algorithms, DPS, in partnership with Arti-Solutions, is able to take the preventative workflows into the predictive world. Assets are often operated in locations where it is not easy to access critical spares which means it often takes longer to restore back to full production rates after a planned or unplanned shutdown. Minimising plant shutdowns is therefore key to improving production. This is where predictive analysis comes in.
DPS has been working with leading academics in the field of predictive analytics and has the ability to tap into the as yet used information provided by an assets real time data streams. Our solutions are focused on:
• Early alert generation and prognosis of off-spec processes and patterns being identified
• Root Cause Analysis, clustering and relation mapping
• Fault diagnosis and predictive fixed & rotating equipment health monitoring
A solution based on these key areas will enhance and deliver operational stability, improve compliance and deliver quantifiable quality improvements. Even the most efficient power plants can benefit from advanced-analytics models. An increasing number of power companies at the outset of their digital journeys are already seeing promising results. A significant number of our clients are experiencing production disruption from equipment failure, poor visibility into operational processes and suboptimal process efficiency. With aging assets comes a declining asset uptime and an increase in operations and maintenance costs. This leads to unplanned downtime which contributes to poor HSE – Q and a decrease in performance standards. With the correct process application DPS and its ecosystem have seen 20-30% improved equipment uptime availability, 30-50% cost reduction by deferring maintenance.
DPS is able to provide an end to end solution to meet our client’s scope of work through an innovative technical ecosystem of world class service and software providers delivering a scalable, open source platform for building and managing the digital twin. Our ecosystem creates a seamless integration between technologies that will ultimately lead to increased productivity, decreased risk and accelerated adoption of new technological developments as synergies between solutions are identified, exploited, and refined.
Historically, digital twin solutions have focused on trying to do everything with a single application or through a single service provider. This has led to many companies being locked into a rigid workflow or hierarchical system that is not agile enough to benefit from a rapidly changing technology marketplace. Utilising an ecosystem, as we have outlined in this document, it allows operators to access cutting edge AI, cloud based 3D visualisation and old school engineering expertise on platform with open APIs enabling them to integrate the deliverables with any data system or data source in a cost effective way without compromise to functionality.