This article, authored by Mark Oosterveer, was originally published in the Vakblad Asset Management (VAM) magazine in July 2025. It has been translated from Dutch to English by Gradyent. Read original here: (NL) VAM - article.pdf
Dutch scaleup Gradyent recently won the Innovation Award at Maintenance NEXT's Innovation expo. With innovative models based on old-fashioned laws of nature, the company optimizes installations that transport energy. An up-to-date asset-base provides a good start in this regard.
Gradyent has grown rapidly in recent years. With physics-based model of energy systems, the company provides insight into energy flows, often heat or more specifically even steam. With that insight, the company also creates opportunities for improving those streams. Over the last year they have been scaling up the experience and knowledge that Gradyent has built up in heat and distribution networks to industry. Because, there too, there is a lot of room for optimization and thus reducing the footprint of companies.
Optimization
The models the company works with are created based on historical data and with the knowns of the hardware used: length, diameter, insulation and medium. These are the properties that determine the behavior of a system. And with the data sets, the models are validated and improved where necessary.
Ann Robin, business development director at Gradyent has over 30 years of experience helping companies in industry with decarbonization. Together with Marco Landwehr, business development associate, she focuses on industry opportunities. "With our model, we bring the energy flows into focus and can properly optimize there. And since we do that, we are also asked to analyze other energy carriers such as fuel gases, CO2, hydrogen and cooling water and improve the corresponding processes. To this end, we have created a real-time physics-based model of the energy systems. That can be used for more purposes. Also to help companies to be compliant with set emission guidelines" says Robin.
Integral
Landwehr emphasises; "About 50% of all energy in the petro- and chemicals industry is stored in the steam systems. You can imagine that optimizing that can have a huge impact. Especially if we tackle the whole process from boilers, CHP and the infrastructure and insulation."
They cite an example of a customer where two boilers were used simultaneously for steam production. By modeling the whole system and using that to control these boilers in real time, efficiency was increased, the balance between the boilers could be shifted, and fuel gases and NOx emissions were reduced. Landwehr; "Better even, we realize savings of up to 4% in gas consumption." The core processes have been optimized at this company, but the utilities have not yet been. And certainly there is a lot to be gained there. Robin; "Certainly those old systems are all heavily over-dimensioned and that's how they are used. By 'understanding' them better and controlling them better, the improvement potential is huge"!
Efficiency
Landwehr further explains the use of models; "It can be an investment to start working with models, for example in conducting a site survey to generate 3D data. And, while it may take an investment to lay the groundwork for collaboration, perhaps there is no 3D data available or missing P&IDs, it is important to mention this is a sustainable investment. This is not a one-time software solution that is ported over after delivery and the relationship ends. Gradyent is a SaaS company and instead looks for longer-term partnerships with industrial companies."
The scaleup thought about that: "First, it ensures that customers' Digital Twins grow with their systems and always remain the most accurate digital copy. Second, we really want to make sure that our solution is actually used by our customers, and by maintaining a long-term relationship in which not only our Digital Twin grows with their system, but also our engineers and developers support the customers during implementation, we can achieve this goal." This is precisely how a Digital Twin with strong return on investment is built in collaboration.
Validate
Robin gives another advantage of an accurate physics-based digital twin. "We already mentioned how having a physics-based, end-to-end, live Digital Twin of the steam system enables users to take a system approach to real- time optimization. But, what often lends additional value to our collaborations is another crucial capability offered by the Digital Twin: validating design plans." That may need some deepening.
Robin: "We see our Digital Twin as being made up of two core aspects: real- time optimization and design validation. Most industrial sites find themselves in the transition to Net Zero: assets are being electrified, certain steam users are being shut down, et cetera. These changes have substantial effects on the operation of the steam system, and estimating their actual effects in advance can be quite a challenge. Will bottlenecks soon develop in my system? Can critical users still meet their steam needs? Is investing in a new heat source really necessary, or can I make do with the two I already have? With the digital twin, which really behaves like the system, we can validate planned changes in advance - after all, we have a faithful model of their steam system, so we can also test the impact of changes to that steam system".
Whereas live optimization focuses primarily on operational OpEx savings, the ability to pre-test the impact of changes delivers significant CapEx value. She hastens to say that in the long run, moreover, this also yields OpEx savings. And proudly, "During our work with Shell, this very design validation aspect of our Digital Twin proved to be of great value."
Understand your system. That could be a nice summary of working with a physics-based digital twin. Especially for the optimization of existing systems, it can help to find the opportunities and to determine in advance whether that opportunity can be realized. And this is exactly how you can accurately steer or reduce emissions, improve efficiency and work on compliance.