Short-term O&M Planning for Offshore Wind Energy

Sammanfattning: Significant breakthroughs are currently being achieved for Offshore Wind Energy (OWE), as offshore wind turbines are moving further away from the shore, benefitting from as vaster wind resource availability. On the other hand, the current trends of the OWE industry make new challenges arise. One of them is the high Operation and Maintenance (O&M) related costs, which are estimated to contribute to around 30% of the total lifetime cost of an OWE project. One of the current solutions to mitigate this high share of costs are Operation and Maintenance Decision Support Models (O&MDSMs) also known as Operation and Maintenance Decision Support Tools (O&MDSTs) when implemented into a software with a user interface. O&MDSMs support decision-making, by informing the decision-maker with insights on the decision at hand. These possess a very diverse array of applications inside the OWE industry, namely wind farm site selection (long-term), vessel composition optimization (medium-term) among others. Additionally, a research gap was found on short-term (days-ahead) O&M planning decision support tools. The present thesis project, developed in partnership with WavEC Offshore Renewables proposes a methodology for an O&M scheduling model (i.e. selecting the best time windows for O&M to take place) that minimizes the wind farm’s O&M operation costs and turbine’s downtime, thus maximizing profitability in the long term of the OWE project. For this, several aspects were considered, namely a set of O&M activities and respective requirements, wind farm expected production and accessibility based on weather forecasts (and their uncertainty) provided by IPMA (Instituto Português do Mar e da Atmosfera), vessel and personnel availability, and a set of generic O&M activities. The wind farm’s energy production estimation considered the wake interaction between turbines, which is a novelty for short-term O&M planning models at the time the present thesis was conducted. Post optimization, a methodology for evaluating the operational risk of each O&M activity incurred during the suggested time windows was also proposed. The suggested methodology is applied to 2 case studies, each containing 4 different scenarios, all of them based on the WindFloat Atlantic wind farm located in Viana do Castelo, Portugal. The obtained results suggest that the proposed O&MDSM methodology can increase the operating profit considerably, given the whole spectrum of feasible scheduling solutions. Given the scenarios considered, the proposed methodology can increase the wind farm operating profit from 7% to over 50%. Moreover, by considering the wake effect, two main advantages were found: Firstly, it influences the optimal O&M activities scheduling decision and secondly provides a more accurate account of the forecasted energy production during the analysed period. Finally, it was also found that for periods with unfavourable wind direction and wind speed which generate an intense wind profile at the wind farm, performing the O&M activities during those intervals may actual be beneficial to the wind farm’s overall production, as the wake generated by a turbine is nullified when it is shut down for maintenance, thus increasing the overall net electricity production.