The current "Subsea Hydrocarbon Leak Detector" sensors are sold by Franatech GmbH to companies in the gas and oil industry worldwide for installation at suitable locations on the seabed in the vicinity of production platforms. The continuously measured values of the dissolved methane content in the water allow conclusions to be drawn about potential leakages on the seabed or on the underwater installations. Since even small changes in the dissolved methane content in the water are detected, the system can identify visually invisible leaks. Due to seasonal fluctuations in water temperature and the geographically varying natural methane outgassing, fixed limit values for the methane dissolved in the water, as currently increasingly used as an indicator for leakages, are not meaningful.
In order to be able to interpret methane measurements directly and independently on the platforms in the future and to derive the associated leakage probability, model-based algorithms are to be developed and bundled in the project on the basis of real data. Finally, the fixed control limits will be replaced by a machine learning based decision support system as a warning signal.