Enhanced Preventive Maintenance of Offshore Wind Farms: Scheduling through Long-Term Hindcast Data Analysis

November 27, 2024

This thesis addresses optimizing preventive maintenance scheduling in offshore wind farms to increase Annual Energy Production (AEP) and reduce downtime losses (DWL). A universal scheduling method was developed to complete yearly maintenance plans while minimizing DWL. Using hindcast data from the Resourcecode database for Europe’s North Sea and Atlantic Coast, "typical years" were computed to describe expected weather windows. Three scheduling methods—historical, adaptive, and hybrid—were developed, incorporating monthly yield thresholds to prioritize low-wind workable days. The adaptive method proved most effective, reducing DWL significantly while completing maintenance, even for larger wind farms with higher maintenance demands. A reference case showed a 44% reduction in preventive maintenance DWL using this approach. Other methods also reduced DWL but struggled in harsher climates or with larger wind farms. A key innovation, the Maintenance Constraint Level (MCL), was introduced to assess scheduling flexibility by accounting for climate and operational needs. Results indicate optimal maintenance conditions around MCL=0.4, enabling a balance between energy production and maintenance needs. Economic evaluation demonstrated up to €0.50/MWh reduction in LCOE in some locations using the adaptive method, highlighting its potential to lower offshore wind energy costs. These findings provide a basis for a decision support tool to improve maintenance planning and optimize offshore wind farm operations.

Authors in the community:

• 

  P

  Paul Henri Claude Philippe Dupin

  ist1109185

Supervisors of this institution:

• 

  R

  Ricardo Balbino Santos Pereira

  ist151719

mechanical-engineering - Mechanical engineering

por - Portuguese

Instituto Superior Técnico