Electromechanical Drivetrain Simulation.

Wind turbines structures are exposed to inclement loading conditions varying from the turbulent wind field to fluctuations in the electric grid. The variation of these conditions, in addition to special events such as emergency stops, has a great impact of the life time of the components. In multi-MW wind turbines, it is common to find a geared drivetrain, which is the interface between the mechanical and electrical domain. Due to the varying conditions, the drivetrain can suffer accelerated damage reducing the target 20 years life of the turbine. This Ph.D. thesis focuses on the implementation of advanced models that consider the electromechanical interaction of the wind turbine structure, namely the main shaft and tower top, along with the gearbox and the generator. This is done with the purpose to advance the integrated analysis of wind turbines; something that is not common until recently. The state-of-the-art in wind turbine simulation is to consider the wind turbine structure with a simplified model of the drivetrain. Therefore, the main purpose of this Ph.D. is to develop a simulation tool capable of estimate the loading in the drivetrain internal components, with special attention to the planet bearings in the planetary stage. In brief, the tool is used for the dynamic analysis of the drive-train components under different loading conditions following certification guidelines. Several numerical simulations demonstrate the capabilities of the tool, and new results show how the lifetime of the bearings are affected by different load cases. The fatigue damage experienced by the planet bearings in the planetary stage is assessed for the normal operation of the wind turbine, by computing the damage equivalent loads for a 20 years period. Several operational modes are identified as the main contributors to the fatigue of the bearings. Second, the ultimate design loads obtained by extreme events such as Low-Voltage Ride through (LVRT), emergency stop and normal stop due to grid loss are investigated. A method to simulate the LVRT based on the grid code requirements from different countries is presented, along with results that highlight the importance of the voltage recovery and its relation to the effect on the bearing loads. Several recommendations are made for the three extreme events in terms of possible load reduction in the bearings. The main goal is to minimize the long-term damage that can be induced by the extreme cases. And finally, reliability analysis using FORM is performed based on two different types of bearing configurations. For this purpose, a bearing stiffness matrix corresponding to each configuration is used in the electromechanical drivetrain simulation tool. Thus, using a parametric study with different dynamic rating i values, it is found that this parameter has an important influence in the reliability, and hence, in the preliminary design of the components. Furthermore, the difference between the damage equivalent loads of both types of bearings is minimal. Therefore, the dynamic rating parameter is found to have higher influence on the bearings reliability. The methods presented in this dissertation can be used to model different drivetrain configurations for preliminary design, based on standard load cases used in wind turbine certification. In addition, it is possible to carry out reliability analysis, which ultimately, is one of the main focus areas when analyzing and designing such complex and costsensitive systems.