Leveraging a digital twin of a reactor cooling system helps maximize production and avoid shutdowns

Simulating automation

Since the 1970s, Practicon has been automating factory processes to help manufacturers optimize production. Over the years, the company has built up expertise in understanding customers’ systems and requirements and designing and implementing ideal solutions to enhance their operations. Practicon aims to always use the latest technology to deliver the best results, and when presented with a new client with particularly complex needs, it was the perfect opportunity to add full multiphysics simulation to its toolkit.

Originally, Practicon’s customer had a manufacturing process with 20 reactors generating a large amount of heat. They managed this using a cooling system that was originally provided by a third-party company, which they had modified and expanded over the years as the plant grew. It was critical to operations but had become unreliable and had limited support, and the business was concerned about the impact if it failed. At the same time, due to the complexity of the system, making changes was not straightforward without the risk of shutting the entire manufacturing process down.

Every 18 months, the company has a two-week maintenance window where the cooling system can be taken offline and changes can be made. Thus, Practicon not only needed to come up with the perfect solution, they also had to ensure it could be implemented within this short timeframe to avoid costing the customer production losses.

“It was clear from the outset that this was different from projects we’d worked on before,” says Martin Jones, director at Practicon. “Once I understood the complexity and criticality involved, I knew we would need the most accurate modeling of the system to ensure success.”

To meet this challenge, Practicon adopted Simcenter™ Amesim™ software, which is part of the Siemens Xcelerator business platform of software, hardware and services, to create a digital twin of the new cooling system to fully test it before implementation.

Validating simulations with mechanical modeling

The existing system suffered from an unreliable control system, a lack of fault tolerance, unstable flow rates and pressures and difficulties meeting operational demands. Before deciding how to overcome these issues, Jones and his team needed a full understanding of the current setup.

Practicon used a mechanical modeling tool for measuring and mapping the existing cooling system and then designing the required changes for a new system design. However, they needed more for this project. “We could simulate the basics of the cooling system, but we couldn’t accurately predict how it would handle any given situation,” explains Jones. “With only two weeks to install the new system, we didn’t have time for any unknowns. We needed a virtual environment that we could test as much as necessary before building anything, such as the effect of reactors being turned on and off, changes in flow rates or pressure or a loss of containment.”

The proposed new solution incorporated the Siemens S7-1500H programmable logic controller (PLC) system. Leveraging its master-slave configuration with redundancy in the network, Practicon ensured that a single point of failure would not bring down the entire system. The system incorporated a ring network to provide resilience against failures, such as an Ethernet cable disconnection. Additionally, the input/output (IO) in the field was also unnecessary to maintain system integrity.

“We had to prove that we could implement an effective solution in the given time frame,” says Jones. “Using our existing mechanical modeling tool gave us a benchmark to compare with when we built the first model with Simcenter Amesim. This gave us and the customer the necessary confidence that the digital twin was an accurate representation of what would happen in the real world.”

Developing an accurate digital twin

Since Simcenter Amesim was new to Practicon, the company reached out to Siemens Digital Industries Software partner Maya HTT for help getting started.

“After the first call, I knew Simcenter was the right tool for us,” says Jones. “Based on our initial conversation, the engineers at Maya HTT quickly created a rough model using the Simcenter Amesim built-in libraries. Immediately, I could see the potential and that it would be crucial to us winning the contract and implementing an effective solution.”

The Simcenter Amesim model was able to run in real time, permitting the test of various scenarios and parameters to ensure the system’s reliability. This real-time capability was crucial in validating the control solution before implementing it on the actual plant.

Engineers from Maya HTT and Practicon worked closely together to develop the model and ensure its accuracy. Once they were confident using it, Practicon could simulate any scenario with the digital twin. “Although it’s a complex solution, we found it intuitive and easy to work with,” says Jones. “Within a matter of hours, our engineers had Simcenter Amesim configured and were running simulations to test the new system design.

“Having built the digital twin of the existing system, we showed the customer that it simulated exactly what was happening in real time. This gave them the confidence that, once we applied our modifications to the digital twin and proved its effectiveness in the simulation, it would work when we implemented it for real.

“To further ensure the reliability of the system, we used the Siemens TIA Portal and the advanced simulator to run the system in a virtual environment. This allowed us to validate the PLC code and mechanical aspects of the solution before implementation, ensuring it met the client’s requirements.”

Proving resilience and passing factory acceptance testing

As part of the factory acceptance test, there were 50 test case scenarios to account for. Using the digital twin, Practicon engineers simulated each of these and generated easy-to-understand charts and graphs that verified the new solution every time.

A significant advantage of the digital twin was the ability to monitor everything that was happening within the cooling system. “Installing physical sensors throughout the system would be impractical and extremely costly,” says Jones. “With Simcenter Amesim, we added virtual sensors to measure flow rates and pressures, anywhere we needed on the entire system. There was no margin for error. If the flow rate dropped and didn’t recover within 300 milliseconds, the reactors would automatically shut off as a precaution. As part of the resiliency of the new solution, we installed a three-pump system that could run on two pumps when needed. Using the digital twin, we proved that the pumps could be taken in and out of operation for maintenance without effecting the flow rate, ensuring that production wouldn’t be interrupted.”

With the new solution in place, Practicon has the digital twin available to run scenario simulations when they need it. “If the customer wants to make changes or has concerns about efficiency or operation, we can run it all through a simulation and show them exactly what’s happening so they can make the most informed decisions,” says Jones.

Creating new business opportunities with further simulation

After the success of the first project, the customer asked Practicon to carry out a similar system upgrade in another section of the factory. It was not identical, so the same approach was taken to model the existing system and provide the effectiveness of the changes with a digital twin.

“This proved to us that the process is repeatable,” says Jones. “Now we know we can use it for other customers too, it opens up new areas of business for us. There have been projects in the past that would have been too risky to take on, but with Simcenter Amesim, we have total confidence in taking them on now. This gives us a significant competitive advantage.”

Having seen the benefits that using Simcenter Amesim brings to Practicon and its customers, Jones is keen to explore the possibilities of more tools from the Simcenter portfolio. “Seeing all the different packages within Siemens Xcelerator has given me ideas for other projects,” says Jones. “We have the potential now to simulate almost any challenge a customer presents us with. I’m sure Simcenter will have a big part to play in helping us win new contracts and successfully complete them.”