Related Articles
Digitalisation Review Partner Insight
Forward article link
Share PDF with colleagues

Revolutionising the downstream

Industrial AI will allow plants to reach their autonomous potential

Reactiveness, resilience and reliability are becoming increasingly important in the refining industry. As a result of fluctuating pricing and demand, the safety, logistics, environmental impact and economics of new scenarios must be analysed daily. Sustaining stable operations and improving competitive advantage will come down to reacting as quickly as possible to real-time analysis while factoring in resilience and reliability.

The ultimate vision for the industry is the self-optimising, autonomous plant—and the increasing deployment of AI across the sector is bringing this closer to reality. However, while refining has been an early adopter of many digital tools, the industry has yet to fully realise the potential of industrial AI.

That is, in no small part, because AI and machine learning are too often looked at in isolation rather than being combined with existing engineering capabilities—models, tools and expertise— to deliver a practical solution that effectively optimises refinery assets.

These are assets that typically rely on engineering models built from the ‘first principles’ of physics and chemistry, which encapsulate key domain knowledge such as process safety and understanding of the industry’s complex systems. 

These models draw on the extensive experience of the world’s best scientists, process engineers and operators. They are highly accurate but have limitations in certain processes. To enhance their accuracy, plant data must be employed to calibrate them to observed plant conditions and performance. Currently, effective model calibration requires considerable expertise and experience.

The solutions supported by hybrid models act as a bridge between the first-principles-focused world of today and the “smart refinery” environment of the future

Building a hybrid model

This is where AI and machine learning have a key role to play. These technologies are rapidly emerging as tools that can greatly accelerate the ability to employ plant data, both to calibrate first-principles models and to quickly create data-based models of phenomena and processes. AI has the potential to lower the expertise required to model process systems, but it must be combined with domain expertise to create the real-world 'guardrails' that make it work safely, reliably and intuitively.

This combination enables what we call 'hybrid models', which effectively bring together AI and first principles to deliver a comprehensive, accurate model more quickly and without requiring significant expertise. And, crucially, they serve as a vital staging post on the way to the self-optimising plant.

Machine learning is used to create the model, leveraging simulation, plant or pilot plant data. The model also uses domain knowledge, including first principles and engineering constraints, to build an enriched model—without requiring the user to have deep process expertise or be an AI expert.

The solutions supported by hybrid models act as a bridge between the first-principles-focused world of today and the 'smart refinery' environment of the future. They are the essential catalyst helping to enable the self-optimising plant.

Many companies today are reaping the rewards of this hybrid modelling approach. Refining and olefin margins are closely related to both plant planners’ and operators’ ability to achieve monthly production that is as close to the plan as possible, and gaps can usually be traced to out-of-date or inaccurate planning models.

One of the largest global refiners projects that the ability to generate up-to-date revisions of these detailed reactor models will deliver value over $10mn annually for a typical 200,000bl/d refinery. This technology is especially timely as refineries contend with dramatic changes in the products they must produce.

Realising the vision

The development of hybrid model solutions will also, for many refiners, be the first step in realising the vision of the self-optimising plant. At AspenTech, we define this as a facility that can automatically adapt and respond to changing operating conditions.

AI has the potential to lower the expertise required to model process systems

Relying on a combination of AI and key domain knowledge, the self-optimising plant will rapidly assess all available data streams, both within an asset and beyond its boundaries. It will rapidly react to changing conditions to achieve the best possible outcome, taking into consideration safety, sustainability, asset health and operational objectives. Furthermore, it will use AI to anticipate future behaviours and provide workers and managers with future alternative operational scenarios.

When the potential of self-optimisation is realised in plants, companies will no longer need to focus on routine engineering and operator tasks but will be able to put all their energies into making faster, smarter business decisions that drive agility. With automation in place, the plant systems will operate with increased efficiency and will be able to react automatically to unanticipated situations. 

It will be some time before self-optimising plants are a reality, but it is this vision of hybrid modelling capabilities that will transform the refining industry.

Antonio Pietri is president and CEO at AspenTech.

This article is taken from our forthcoming Digitalisation Review, which will be published in November.

You can hear more from Antonio in our webcast, PE Live: Digital Business Transformation. Click here to register

Also in this section
Liberian Registry hits out at proposed EU ETS shipping extension
15 October 2020
The world’s second-largest vessel registering service opposes what it sees as European overreach
Regulation to play a key role in India’s gas ambitions
15 October 2020
The US experience suggests building pipelines is not enough. An effective regulatory regime is also required
Apicorp sees more woes for Egyptian and Algerian LNG
13 October 2020
The multilateral lender flags risks of ongoing shut-ins and difficult contract negotiations