Enhanced Electrical Signature Analysis (e-ESA) for Offshore Wind Permanent Magnet Generators

Offshore wind turbines can generate large amount of renewable energy, but the operation and maintenance (O&M) can be challenging due to the harsh and not easily accessible environment, which can increase operational costs. This work aims to develop methodologies that use the electrical signals of the power converters to detect faults that are not readily detected by existing protection systems of the generator. Currently there are condition monitoring systems based on vibration, thermal and electrical sensors but they can't detect faults like turn shorts and increased crimp resistance in direct-drive permanent magnet generators (PMGs). Detecting these faults early before they evolve into catastrophic failures can help to optimize O&M operations, prevent unplanned downtime and therefore reduce the operational expenditures (OPEX). The detection methodology uses the harmonic content of the electrical signals. The case study of a 6 MW turbine presents simulation results under healthy and unhealthy conditions, and the field data collected from operational wind farms. The results of the features considered show a promising separability between healthy and unhealthy conditions, which indicates that this methodology could be implemented for condition monitoring of PMGs in offshore wind farms.