PredictCUI: Prediction of water liquid and vapour migration for mitigating corrosion under insulation

Insulated pipes are found in places where hot gas or liquid needs to be moved from one place to another place nearby, whether this is the hot water in your house or large steam pipes in industrial plants. The insulation, which sits like a woolly fleece around the pipe wall, helps to reduce heat loss, costs and emissions. When insulated pipes are found outdoors, they need a raincoat just like we do, made of a thin metal layer called cladding. However if this raincoat has a hole, the consequences are much worse than a wet shoulder. If the cladding on an insulated pipe is damaged, the water leaking in can cause a major problem known as corrosion under insulation (CUI). CUI is corrosion damage to the pipe carrying hot fluids. In the worst case, CUI can lead to major accidents if a pipe carrying flammable or explosive fluids is damaged. Avoiding CUI is such an important issue in petroleum and chemical industries that the total maintenance cost due to CUI in Norway is more than 1 billion NOK per year. Part of the reason for this big cost is that operators cannot predict where CUI has happened, so hundreds of kilometres of insulated piping must be manually inspected on a regular schedule. The PredictCUI project will enable development of a humidity monitoring system, by closing knowledge gaps in the understanding of humidity transport in pipe insulation. That's because by measuring humidity levels inside the insulation, it is possible to locate where water has entered the system. Combining the new fundamental knowledge on humidity transport that will be gained from theory and laboratory work, with the experience and computer models developed in PredictCUI, will help bring sensor-based CUI monitoring solutions closer to reality, helping to reduce costs and risks in affected industries. PredictCUI is a multidisciplinary research collaboration between SINTEF, NTNU and TU Eindhoven, and is supported by partners Shell, Equinor, Gassco as well as the Research Council of Norway. In 2020 the project has started up and work has begun on building and placing out sensors to measure moisture in insulation, the laboratory investigations of water transport in insulation have begun, we have initiated the computational modelling of the moisture migration, and a thorough mapping has been performed of the existing standards and documents governing how insulation is used and encapsulated today, as well as how the risk of CUI is handled in the current situation where sensor data and models are not available. In 2021 the project has had its first full year of normal operations, where new moisture sensors have been developed, deployed and used to collect data from the field, laboratory measurements have been performed on vertical insulated piping, we have developed more advanced modelling tools for moisture migration in insulation that ensure thermodynamic consistency, and test have been performed to ascertain how modern centimeter-accuracy positioning technologies can be employed in CUI monitoring on industrial facilities. The two PhD students in the project have both come off to good starts on their work in the project. In 2022, the project had full activity in all work packages, and work has been proceeding on laboratory experiments on vertical pipes, development of numerical models for moisture transport, and analysis of data from moisture sensors deployed in the field. Results from the project led to the submission of several journal papers this year, and a popular science article was published in Teknisk Ukeblad. The project also contributed to hosting a "CUI Open Day" at the Energy Transition Campus Amsterdam, where over 50 representatives from the industry participated to learn more about predictive CUI maintenance and the results obtained so far. In 2023, the project had its last year of full activity in all work packages. Several new papers from the results obtained were submitted for peer review, and were presented at workshops and conferences. Special emphasis was placed on development of numerical models for moisture transport in insulation, and algorithms for data analysis from moisture sensors deployed in insulation. These have been developed to become powerful tools during 2023. SINTEF is now working on developing the technology further, and is considering commercial development of the results. 2024 was the last year of the PredictCUI project, and the activity was centered around the final work of the two PhD students. Of these two, HyeJeong Cheon defended her thesis in september, and Hristina Dragovic travelled to TU/Eindhoven to perform the final measurements of moisture transport in insulation using quantitative NMR techniques, collaborating with Dr. Leo Pel. The thesis defense for Ms. Dragovic is anticipated in 2025. Several of the results from PredictCUI were published in leading scientific journals in 2024.

PredictCUI has accelerated the development and deployment of CUI monitoring using spot sensors that measure humidity and temperature in the insulation. The physical understanding of water migration in insulation has been developed and disseminated in the project, such that researchers, vendors and end users all have an increased understanding of how to apply spot sensors and make use of the results. Currently, several thousand sensors are deployed commercially, with Norwegian vendors holding a large portion of this growing market. The industrial partners in PredictCUI are driving the technology further in their internal R&D and in collaboration with the PredictCUI team. Future work is required on maturing the concept from the end-user perspective, both to build further confidence that enables use of sensors in a true retrofit fashion, to integrate CUI monitoring into the ongoing digitialisation of assets, and last but not least, to adapt end users' CUI inspection and maintenance workflows such that the full potential of the technology is realised.

Corrosion under insulation (CUI) receives special attention in the oil and gas industry, specifically off-shore, as consequences of spillage or pipe rupture can be catastrophic. CUI is a matter of grave concern because it may break the structural integrity of insulated pipe networks. Corrosion rates can be as high as 20% of pipe wall thickness per year, demanding exhaustive maintenance work, consuming up to 50% of the preventive maintenance budget. PredictCUI will initiate a shift towards predictive CUI maintenance. Targeted maintenance will mitigate the risk of major accidents on the Norwegian Continental Shelf (NCS), support improvements in efficiency, and increase health and safety of maintenance personnel. The security of supply and competitiveness of the Norwegian petroleum industry will be strengthened. PredictCUI will enable the development of a humidity monitoring system by closing knowledge gaps in the understanding of humidity transport in pipe insulation. A coupled multi-scale fluid dynamic-thermodynamic computational model will be developed and validated against new laboratory setups to measure humidity transport. Robust, retrofittable, long lasting, wired and wireless sensors will be tested in close collaboration with Equinor, Shell and Gassco, under realistic laboratory and on-site field conditions. New knowledge obtained in PredictCUI will contribute to lift the TRL of sensor-based CUI monitoring solutions towards proof-of-principle and lay the groundwork for future standardisation and commercialisation. PredictCUI will enable data integration and accelerate offshore digitalisation by developing a scalable and reliable model-sensor interface prototype. Findings from PredictCUI are relevant throughout a range of process application industries. PredictCUI will be a multidisciplinary research collaboration among SINTEF Energy Research, SINTEF Digital, NTNU PoreLab, NTNU Department of Energy and Process Engineering and Technical University of Eindhoven.