The European Union’s policies on energy and the environment are aimed at promoting and developing offshore wind platforms. This means that the Spanish electricity system is going to depend increasingly on this type of electricity generation system. The wind turbines for this case are larger and more complex, and require high standards of safety, reliability, availability and maintainability. This project approaches this problem with the ultimate goal of comprehensive and optimal management of this type of wind turbine farm.

The project starts with the aim of extending and completing the National WindSeaEnergy (DPI2012-31579) and OptiWindSeaPower (DPI2015-67264-P) projects, where the most important scientific references in high impact journals were analysed, observing that there are great deficiencies in the mathematical models that allow the analysis of the signals that are being monitored to determine the state of the structures, as well as in the optimal management of the wind turbines and wind farms

ProSeaWind intends to continue and complete this study initiated in the field of the rotating and structural elements of the wind turbine, and the management of their maintenance, making a more exhaustive study of the monitoring systems and signal processing methods for the elements of this equipment. The data obtained from the aforementioned projects will be taken as a reference, in addition to the European projects OPTIMUS, NIMO, WINPRO and the National IcingBlades. A life cycle cost model will be developed for the predictive maintenance system. It is proposed to use mathematical models based on the analysis in time, frequency and time/frequency, as well as Transformed Wavellets, Neural Networks / Artificial Intelligence, methods based on the extraction of characteristics of the signal and derived from the Transfer Function of the system. The multivariate analysis will be carried out by means of Logical Decision Trees, which will be analysed using Binary Decision Diagrams, and the measurements of importance created by heuristic methods. This will allow the condition of a wind turbine to be fully controlled and optimised. For the optimal management of the offshore wind farm it will be raised as a Markovian Decision problem, and the Restless Bandit indices will be analysed in order to determine the structure of the Whittle-type indices in different contexts. Finally, new cost-based significance indices will be created and the optimization problem and its resolution will be formulated through metaheuristic methods to determine the optimal investment policy in the management of this type of wind farm.