Digital twins are quickly becoming the place where plant engineers, site personnel, maintenance crews augmented by automation, and AI capabilities converge across multiple systems to improve the ways they work, driving toward their ambitions of smarter, safer and greener operations.
Successful technology advancement and adoption are essential to the energy transition. Traditional sources of energy are being challenged by calls for increased production at lower costs. The number of new energy assets to be managed is increasing, with overall renewable energy capacity additions globally growing by nearly 13% in 2022 and permitting for additional renewable energy developments becoming a key policy focus across Europe. In the United States alone, the backlog of large-scale electric generation and storage projects applying for connection to the grid reached 5,000 by the end of 2020.
The energy system is also more distributed and decentralized due to limited land availability and regulatory obstacles like distance rules, biodiversity concerns, and costs. Even when land is accessible, strategic placement of developments becomes challenging.
Renewable energy sources must be linked to aging infrastructure, the demand for CCS is growing (130 commercial-scale CO2 capture projects were announced in 2021), and volatility in the global energy markets adds complexity to an industry that is moving toward electrification. The speed of acceleration required to handle this growingly complex energy system hinges on the industry’s ability to continuously adapt, evolve, and integrate new ways of working – with digital tools taking center stage. Generative AI has over the past few months become a commonplace topic of discussion. How might Generative AI help address these challenges?
To understand what generative AI can do for the energy industry, let’s take a step back and look at digital tools and the importance of a data foundation. There are innumerable data points across the value chain. Here’s an example: a major plant turnaround requires input from engineering specifications, current-state plant conditions, work order schedules, crew availability information, project plans and so on. It would be a stretch to claim that all this data is readily available and easy to use – but the potential of this data is huge. We are looking at a reality where key information is distributed across systems, data ontologies and models that are, more often than not, completely disconnected.
Most operators have vast amounts of data. These data points are both costly and valuable – in fact, the oil and gas data management market alone is projected to reach over $90 billion USD by 2030. The volume of data is not the problem – making it useful, however, is a different story. At a single operating facility or onboard a single maritime vessel, data typically exists in different formats, data models and semantic models – and different standards are often applied based on local preferences, acquisition history or another overriding factor. With this challenge at hand, what is the most effective approach to bring these together in a meaningful way and unlock that potential value? A digital twin.
For a single piece or type of equipment, there can be hundreds of data points...
... and each facility or vessel can host thousands of pieces of equipment.
That means potentially millions of data points that need to be brought together!
The Industrial Work Surface puts these data points in context. Information is collected, sorted, and displayed in a single unified environment.
From documents and drawings to reports, real-time sensor outputs, time series historians, P&IDs, and more, digital twins offer a golden opportunity to bring these together in an integrated environment for complete data contextualisation. The first step is data retrieval and ingestion to make this information available in a cloud-based environment. From here, work planning and surveillance at the grassroots level can be initiated as first steps, followed by scenario planning and predictions that start moving toward more autonomous operations.
Once this solid data foundation is in place and a digital twin is populated, it becomes possible to add AI, dynamic simulations, and IoT devices like sensors. Of course, this means increasing requirements for data integration as the management of individual and even entire portfolios of assets scales to incorporate an increasing number of users and systems.
That’s why data, and technology that makes data useful, lies at the core of the energy transition. The right technology unlocks the full range of asset performance data, in a way that makes sense to users and stakeholders. It provides an intuitive and actionable way for people to access and use information, from anywhere in the world. It forms the foundation for business owners and operators to pivot from driving incremental production optimizations to incorporating AI that sees increased ROI, improved energy efficiency, and beyond.
Once a digital twin is up and running (powered by that solid data foundation, of course) industry operators are equipped to deal with the reality of rethinking asset and vessel management strategies for the future. They are also well-positioned to leverage the opportunities presented by the tech-driven energy transition while balancing requirements for more environmentally conscious operations. By having data available and mapping business needs to workflows and use cases where the digital twin can bring value that scales, businesses can witness the true strength of a digital context.
Example of industrial workflows that are enabled within a digital context
With a solid data foundation presented in a digital twin where users can plan, execute, and close out entire workflows from start to finish, a fully digital context is born. There are new possibilities for remote surveillance, support, and control. For example, a turnaround can be planned in a digital setting by multiple teams located across the world. Logistics and resources can be tracked in the same environment where planning takes place, experts can weigh in remotely, and field workers are armed with digital tools that give full access to asset data – historical, current, and upcoming – at any time. And with AI capabilities and simulation, plans and solutions can be rapidly tested, deployed, and refined at the click of a button.
When a combination of data standards, fuzzy rule matching, and a powerful data graph are present, a digital twin is well-appointed to deal with the myriad of complications like different naming conventions, indirect references, and misspellings that arise when it comes to data ingestion and contextualization. However, there are always outliers, like data that might be missing or misplaced. Based on experience and knowledge of a particular facility, a human operator may be able to find the connection and fix the outlier, but programming a rule to catch these is more challenging. That’s exactly where generative AI comes in. By using natural language processing on human-readable text – for example, found in the description of a tag – and matching this to examples found by queries, your digital twin can begin to suggest automatic proposals for fixing this data. Over time, this can help repair the dataset for the operator in a semi-automatic fashion. And when metadata is insufficient or unstructured datais too complex, Generative AI and natural language algorithms can extract this information in mere seconds.
Generative AI and natural language will not only benefit data contextualisation but also make great strides in improving the human-technology experience. Possible use cases include:
The tech-driven energy transition holds the powerful potential to lower repair costs and minimize emissions, increase production, improve drilling efficiency, and limit equipment downtime – all the things that operators worldwide strive to achieve as pressure mounts for near-net-zero operations and working environments that are smarter and safer than ever before. Generative AI is another tool in the toolbox to help us make the best decisions for people and the planet.
At this early stage in the Generative AI journey, the full capability of the technology is still uncharted. But use cases for energy and maritime are already being tested – Generative AI has joined the industrial transformation journey. It will play an increasingly vital role in ensuring the seamless availability of business data for informed decision-making and efficient work execution. And as it continues to be integrated into existing digital twin technology, it is well poised to have a substantial material impact on existing and new ways of working, ushering in new dimensions of efficiency, reliability, and sustainability in industrial processes.
The Industrial Work Surface is an end-to-end dynamic digital twin ecosystem where end users are at the center of intelligent assets, perfectly positioned to access the information they need. Get in touch to see what our AI-infused Industrial Work Surface can do for you and your business, today and in the future.
