Computer-implemented method for optimizing the operation of a drivetrain of a wind turbine

A computer-implemented method for training a classifier for failure diagnosis of a drivetrain of a wind turbine, comprising: training a physics-based model (100) of a drivetrain of a wind turbine with operation data (151, 152) of the drivetrain in normality conditions, thus providing a normality hybrid model (200) of the drivetrain of the wind turbine; modelling at least one anomaly situation in the normality hybrid model (200) and training it with operation data (151, 152') of the drivetrain in failure conditions, thus providing a fault hybrid model (500) of the drivetrain of the wind turbine; applying a set of input data of the drivetrain in normality conditions (151) to the normality hybrid model (200), thus obtaining a set of synthetic data in normality conditions (280) for a certain parameter of the drivetrain; applying a set of failure data (150") and a set of synthetic input data (710) associated to the at least one anomaly situation of the drivetrain to the fault hybrid model (500), thus generating a set of synthetic failure data (750) for the at least one anomaly situation; from a set of output data of the drivetrain (152, 152'), the obtained set of synthetic data in normality conditions (280) and the set of synthetic failure data (750) for the at least one anomaly situation, obtaining a set of condition indicators (8105) for the at least one anomaly situation; training a classifier (900) with the set of condition indicators (8105) for the at least one anomaly situation. A computer-implemented method for optimizing the operation of a drivetrain of a wind turbine, comprising using the classifier (900) as previously trained.