Among the frontrunners
He points out that digital twins are already being used in many industries, such as construction and energy, where they support operations, maintenance, and energy optimization, among other things. However, their use in wastewater treatment is relatively new.
Novonesis is among the companies at the forefront of using digital twins for the biological treatment of industrial wastewater. The company is responsible for operating the joint industrial wastewater treatment plant in Kalundborg, where wastewater and biomass from both its own and Novo Nordisk’s biotechnology and pharmaceutical production are treated.
It is a complex treatment process in which the wastewater composition varies, and many biological processes occur simultaneously, including microorganisms breaking down organic matter and removing nitrogen and phosphorus from the wastewater.
“The pressure on industrial wastewater treatment plants means that wastewater must be treated more effectively to reduce the discharge of organic matter and nutrients. At the same time, increasing volumes are further straining facilities. The treatment process is too complex for process control to rely on gut feelings or minor adjustments. The many biological processes influence one another, and a change in one place can cause a reaction in a completely different system. It is therefore no longer just a matter of reacting to problems, but of being able to anticipate them and maintain control over the entire process. This is where the digital twin comes into play,” says Krist V. Gernaey.
Solid decision-making basis
The digital twin is used to optimize the wastewater treatment plant treating wastewater from Novonesis and Novo Nordisk’s production facilities in Kalundborg. For many years, he and his research group have been developing mathematical models that can simulate and optimize biological and industrial processes.
The models provide a detailed understanding of what happens at a wastewater treatment plant—from how the water is treated and how bacteria work, to how energy and chemicals are used in the treatment process. The models have been tested in collaboration with Danish utility companies and as part of several industrial research projects.
Over the past ten years, DTU—in collaboration with Novonesis—has used these models to develop a digital twin of the companies’ industrial treatment plants in order to gain a better understanding of their processes. In practice, Novonesis’ digital twin is a software-based system that runs on a server and is continuously updated with data from sensors measuring temperature, pH, wastewater volumes flowing through the plant, and many other complex variables. The Danish engineering firm DHI has built the solution into a software platform that is directly linked to the plant’s data.
With the combination of mathematical models and real-time data, it is possible to calculate and predict what will happen at the plant, which offers a solid decision-making basis.
“We expect the digital twin to provide us with insights into what is happening in the biological part of the facility—insights that we cannot physically observe using measuring instruments. It also serves as a testbed for testing operational changes, different configurations, and control scenarios without affecting day-to-day operations. It provides a clear picture of the implications in terms of environmental impact, CO₂ emissions, and cost,” says Sille Bendix Larsen, Environmental Technology Specialist at Novonesis.
Treatment plant control is important
Today, Novonesis is using the digital twin to, among other things, simulate how the existing control system can reduce nitrous oxide emissions from an active sludge plant. They also use the technology to assess the impact of investment projects on capacity and emissions, and to analyse how new raw materials affect the treatment plant. The model has also been used to redesign the plant to better capture resources in the wastewater. The next step is to fully integrate the digital twin into day-to-day operations.
In the long term, Sille Bendix Larsen hopes that the digital twin will be able to control parts of operations and send signals to the plant’s control system as needed. As a result, it can both support decision-making and serve as a practical assistant to operators—for example, during troubleshooting—while also helping to identify risks and plan maintenance more systematically. That says something about how important the control of the plant is, Krist V. Gernaey explains:
“Managing their own wastewater and waste has become a fundamental part of companies’ responsibilities. When you produce something, there are also by-products that need to be cleaned and treated. The treatment plant is therefore not something that is separate from production—but rather an essential part of the very production process. Without the facility, the wastewater would strain the rest of the wastewater system and could ultimately lead to pollution of the aquatic environment. At the same time, we’re working to reduce energy consumption and make better use of resources.”
As an example, Krist V. Gernaey mentions that residual products from sludge can be utilized and used as fertilizer. In some cases, the treated wastewater from Novonesis and Novo Nordisk can also be reused in production, reducing the need for fresh water.
At the same time, both water and resources from wastewater—including biomass—can be utilized at Kalundborg Symbiose. Here, businesses and utility companies are collaborating on recycling resources such as wastewater, treated wastewater, used cooling water, biomass, fertilizer, and biomethane. One of the resources that is also utilized is the heat in the wastewater. Kalundborg Forsyning (utility) receives a portion of the treated wastewater and recovers the heat using a heat pump, which transfers it to the district heating network.
