A Digital Twin is More than Just a Clone

Cartoon lego men digital twin

A Digital Twin is More than Just a Clone

A digital twin of an operating facility is more than just a clone. It’s a unique and valuable asset in its own right.

Genetic Material for a Digital Twin

Several years ago, a large liquefied natural gas (LNG) project in Australia (at the time one of the top 10 largest oil and gas developments on the planet), divided the build into four engineering contracts (wells, gas plants, pipeline and LNG plant).

At the conclusion of the build, the owner had a good view of how each asset was built (the engineering data), but the data was asset specific, and in different systems.

They then had to build a completely separate view of how the assets would work (the operating system). They had yet another system (ERP) that told them what the assets cost to maintain, run and staff. The maintenance team had another system that tagged all the assets, issued workorders and maintained maintenance records. And the Finance team built models to try to predict the future economic performance of the assets under different scenarios.

They had not so much of a digital twin of their new business—more like a collection of unorganised digital genetic material.

Oil and gas is a data-rich and asset-rich business. Oil and gas wells, plants, pipelines, refineries, tanks, vessels comprise the assets, and a key goal of the operators of these assets is to keep the assets running at peak performance at low a cost as possible. A lot of resource businesses are the same – mining, mineral processing, pulp and paper, farming.

Asset builders generate enormous quantities of data about these assets when they’re built – who made them, the nature of the materials used, all the parts, the warranties, the costs, the expected performance profile. Multiple engineering disciplines contribute to the build – mechanical, electrical, structural – each with their own language in terms of diagramming standards, taxonomies and practices. This data is the raw genetic material for a digital twin. 

There typically hasn’t been a lot of thought given towards the future value of this data when it’s created, so it’s not unusual to find it spread out across lots of different proprietary systems.

Organising this genetic material is the first step towards creating a true digital twin.

The First Digital Twin

The idea of a digital twin is not a foreign idea in oil and gas. I think the first examples of creating a digital version of an asset originated with the geology team. They are immersed in the world of 2D, 3D and now 4D seismic, well logs, simulations and resource models. After all, it’s not like you can simply take an elevator down to the reservoir and pay a courtesy call to inspect the holdings. Yet we still manage to find the stuff, and financial markets happily place a value on the presumed existence of these recoverable resources.

Clearly, geologists must satisfy themselves with building an understanding of what lies below through the information that they have about the geology. They build a digital twin of the subsurface and simulate how it might behave under different scenarios (such as drilling and fracking programs). It’s not a stretch to think that  oil and gas quite probably invented the idea of a digital twin.

Digital Twin 2.0

There’s now enough technology horsepower available at relatively low cost, that any business, including those as complex as oil and gas, power plants and renewable energy farms, even full value chains, can build a fully functioning end-to-end digital twin of their business, and not just the digital twins of the specific assets that make up that business.

The digital building blocks that make this possible are cloud computing (the huge data centers and vast data storage devices), clever mathematics, inexpensive sensors, and new programming platforms for building the digital twin.

A fully functioning digital twin of a business includes many layers of data that work together to provide a rich, fully integrated and analytically deep software version of the business:

  • The engineering content (diagrams, specifications, configurations) that describe the physical asset in digital terms for the engineering disciplines
  • The physical constraints of the various assets (their operating capacities, throughputs and pressures) that bound how the asset can physically behave
  • The operating parameters of the assets (input energies, consumables, byproducts and emissions) which bound the asset’s performance
  • The financial description of the assets (fixed build cost, operating cost per unit) that yield the economics of the business
  • The uncertain elements (customer demand, weather events, supply disruption) that are the real world conditions with which the business must cope

The digital twin can have any number of these kinds of variables, whatever makes the most sense for the asset owner. For example, a digital twin of a wind farm would want to include the wind intensity over time, including day/night and seasons, and the demand for power. An oil refinery would want to include the crude slate with its variances of TAN values, sulphur content and heavy metals, its vessel arrival rate and market demand. A tank farm would want to include customer orders, supplier shipments and blending opportunities. A mine would want to include its shovels and trucks and their behaviour.

The Digital Mimic

Most companies content themselves with a digital mimic. The mimic, like a parrot that mimics a human voice, cannot do all the things that the real thing can do. An industrial mimic might integrate engineering diagrams with the ERP system (a good thing, and very clever), but not provide a way to visualise the asset and how it physically works under different conditions.

You can tell if you have a digital mimic if you only “model” your assets and your business using Excel. A model is just a clever mimic. A true digital twin can simulate in the fullest sense how assets and business behave under different conditions.  

The value of the digital twin

Here’s just a few of the use cases available to oil and gas companies who have digital twins of their business assets.

Reveal flawed assumptions

Many businesses run on sets of operating assumptions, handed down from worker to worker, that may have been accurate when the business was first created, but may no longer be valid. The digital twin can validate key assumptions and reveal which ones are no longer reliable. This is very valuable at times of abrupt supply and demand changes, as we are experiencing in the pandemic. 

Extract more value

The digital twin can reveal hidden value opportunities, by improving asset utilisation of key assets, running assets fully, or avoiding certain high cost ways of operating the assets. In times of capital constraint, improving the turnover of existing assets may be the only available way to improve the business. 

Make capital trade offs

A digital twin reveals bottlenecks and capacity constraints in the business as it operates. These are often not plainly apparent through the financial results. A good quality digital twin allows you to add, enhance or take away assets from the business and simulate the business to check out its behaviour, and to inform capital choices.

In a tank farm setting, should another tank be added, or should a line be built between two unconnected tanks? In a wind farm example, should management add a lower cost turbine with high maintenance costs that operates well at low wind load, or a high cost turbine with low maintenance costs that operates reliably under very heavy load?

Sharpen up planning

Sometimes you need to look carefully for the opportunity to create a digital twin. For example, oil and gas well planning, which concerns resource optimisation (targeting the reservoir sweet spots), reservoir characteristics (the mix of liquids being produced), actual well placement, surface facilities construction, and market conditions is spectacularly hard modelling but dead easy for a digital twin. 

Identify key variables

The digital twin of a tank farm, for example, can help identify which customers are most valuable and which ones incur the greatest cost. Operators can tune pricing to protect their best customers. I worked on a game theory project that also helped identify key customers, but not over time, and not during operations.

Building your digital twin

How might your organisation go about building the digital twin?

Strategize first

There are a lot of choices to be made in delivery of the digital twin. Where do you start?

Should you trial a single discipline (mechanical), on a single asset, to build that first mimic to figure out the value? Or do you aim to produce the digital twin on the next greenfield asset? Do you concentrate on cleaning up existing data (using clever digital solutions!), anticipating an eventual move to a new digital platform? Or all of the above? How much value do you think you can capture using your digital twin?

Meet some industry leaders

The vendors who sell the digital twin solutions (Stream Systems, for example) can direct you to the early adopters and the outcomes they have experienced. The early adopters can also help you avoid repeating common errors.

Consider a digital mimic

Jumping directly to a digital twin is tough if you’re not organized for data sharing, the systems are disparate, and no one trusts the data. A good safe move is to build the digital mimic and take it from there. 

(The first version of this article appeared on June 5, 2017).

***

Check out my book, ‘Bits, Bytes, and Barrels: The Digital Transformation of Oil and Gas’, available on Amazon and other on-line bookshops.

Take Digital Oil and Gas, the one-day on-line digital oil and gas awareness course.

Mobile: ☎️ +1(587)830-6900
email: 📧 geoff@geoffreycann.com
website: 🖥 geoffreycann.com
LinkedIn: 🔵 www.linkedin.com/in/training-digital-oil-gas

 

No Comments

Post A Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.