Given the increasing level of technology readiness associated with subsea processing equipment under the trust of ambitious and challenging subsea field development schemes, in order to allow or enhance the hydrocarbon production and/or to reduce topside space requirements, oil and gas Operators are more keen to sanction new projects involving active subsea equipment.
The above scenario will most likely be the driver for a step-up towards a further communication link between subsea equipment (e.g. subsea pumps, compressors, separators, ancillary systems, etc.) and the end user. This link, enabled by a brand-new Condition Monitoring System, will be in addition to already consolidated subsea control system technologies and will pave the way for predictive maintenance operations of subsea processing systems.
Through the acquisition of subsea processing systems specific parameters detected during their operation and comparison with the relevant normal (baseline) conditions associated with each piece of equipment and/or typical component failure modes, it will be possible to monitor the deterioration of the equipment. It will also be possible to understand if it is time to plan a maintenance campaign for the system or if it can be postponed or moved up with respect to the MTBM of the system.
The condition monitoring system will collect a huge amount of data regarding system and components status, operating conditions and performances, and by tracking the collected information, it will be able to build a system history. This becomes the feed in a machine learning process that will regularly increase the level of confidence under real conditions of the subsea equipment laid far from the Operator's sight.
This paper will focus on a condition monitoring system concept applied to Saipem technologies for subsea processing.
