New modelling concept combining system description, phase dependent failure and restore durations and operation schedules at multiple levels.

The answer lies in the methods and tools that are developed to ensure the reliable, efficient and affordable operation of such a mind-bogglingly complex technical infrastructure. The same tools apply in natural gas infrastructures that supply heat to millions of houses.

To meet these challenges, CERN recently signed an R&D agreement with Ramentor Oy , a high-tech company based in Tampere, Finland. In a three-year project, engineers from CERN, Ramentor and other institutes participating in the Future Circular Collider study will develop a tool to assess the effectiveness of different designs and operation schedules for large technical systems.

CERN and Ramentor developed a new concept. It integrates industry best-practises in modelling of complex systems and the concept of variable durations for time to fail and time to restore that takes into account the current condition and history of a facility, based on probabilistic simulations. The new tool allows defining operation schedules at different levels ranging from seconds or months up to the entire lifespan of an installation.

More importantly, the system lets the engineers define production functions at any system level and with limited programming skills thanks to the development of a smart user-friendly interface. These functions are the key indicators used for the performance assessment. While an industrial plant produces goods, the chain of accelerators produces particle collisions. From an engineering perspective the similarity is striking. A sensitivity analysis module lets engineers as well as persons in charge of the facility’s financial performance perform a what-if analysis.

A Result of Extensive Research

Today, the simulation of a ten year LHC operation still takes tens of minutes. First trials to divide the problem into manageable chunks for a cloud computing environment that can expand with the size of the system have been performed. They have demonstrated that the final set can lead to an interactive experience: users could check almost instantaneously how the performance and operation cost of a system changes when it is modified.

This tool will be used to create an operation model of the LHC and its injector chain that can be validated using actual LHC operation data. This valuable model will directly be used as additional aid to verify the operation concepts for the High-Luminosity LHC upgrade. It will also be used by the FCC collaborators to develop and commonly refine system behaviour models and explore different sets of operation parameters by using a cloud setup.

There is a good understanding of the needs and the next steps in the development of this tool. Perhaps it should be noted that a prototype has been built in a pre-curser project with the company over the last year. The new modelling concept combines system description, phase dependent failure and restore durations and operation schedules at multiple levels.

The project also has significant training value. Two doctoral students are actively contributing to this endeavour, while more research opportunities will arise. The development of new method and this innovative tool is also part of a reliability training at CERN, which takes place two times per year.

Ramentor’s cooperation with CERN have not remained unremarked given the wider interest of many industries in RAMS engineering. Wärtsilä , a Finnish energy systems producer and energy supplier, with 18’000 employees worldwide, has recently decided to make their natural gas-based energy production more efficient and reliable. The company is looking how to develop a new concept for a turn-key natural gas terminal and became interested in the modelling, operation simulation and efficiency improvement assessment capabilities that the new tool chain for the FCC study may offer.

Research in the field of RAMS also creates benefits for similar large-scale research infrastructures including particle accelerators, telescope arrays or manufacturing plants. The creation of open model and system specifications will enhance the sharing of knowledge in our community. The new software will be accessible to CERN for its scientific project and study activities.

Text: Johannes Gutleber​ (CERN)