Comparison of Different Model Depths of a Direct Drive Wind Turbine Regarding the Generator Air Gap Sensitivity

The goal of this master thesis is to compare different model depths of a direct drive wind turbine regarding the generator air gap sensitivity under the gravity’s impact, the magnetic pull and the wind loads in the FEM and MBS. The wind turbine that has been analysed has a rated power of 3.6MW and it works with a nominal speed of 12rpm. The rotor’s diameter is 130m, the hub’s height is 115m and between the ground and the rotor axis there is an angle of 5 degrees. All the parts of the generator are in steel. The mass of the generator is 160tons and the mass of each blade is 27tons. The air gap has a length of 1.3m, a diameter of 5m, and a size of 5mm. The first model that has been done is the FEM one. At first, we need to do the convergence analysis for the mesh. The next stage has been made with the generator MBS model. It has been made with only two bodies, one for the stator part and one for the rotor part. In both models the stiffness of the bearings has been applied and the auto MAC matrix and the MAC matrix have performed. The results from the MAC matrices show that the behaviour of the eigenmode of the FEM and MBS models has a good matching with some coupling eigenmodes. After the validation, the impact of the loads was analysed. The static analysis has led to an air gap displacement, under the effect of the only gravity, in the range between -440µm and 370µm. Different models’ depths for the magnetic pull are compared and the results show that the air gap displacement is always in a range from -534µm to 462µm. The second model depth analyses how much the bearings surface on the pin and on the connector must be. These simulations include also the wind loads and a displacement always between -1081µm and 1062µm. Moreover, the impact of the stiffness of the bearings has been estimated and one simulation with rigid bearings has been done. The maximum air gap displacement found is of 382µm. This means that the biggest impact on the air gap size is due to this. The MBS generator has been analysed for the same load cases, but the air gap size has been studied only as the displacement between two markers in the middle of the air gap: one constrained with the stator and one constrained with the rotor. In order to see how many eigenmode (without the IRM frequency) are needed, different simulations have been run. To have a good matching between the FEM and MBS output 80 eigenmodes are needed. Before implementing the generator inside the turbine’s final model, it was necessary to identify the number of eigenmodes and the IRM cutoff frequency, with two reliable results: 10 and 400Hz. The Campbell diagram for the wind turbine was drawn and it shows that there are 6 resonance points in the working range, but there is no resonance point at the nominal speed and there is also no resonance point for the 3p excitation. This means that there are not dangerous criticalities. Only three different models’ depths for the magnetic pull have been analysed and the results show that the difference between them is not relevant (1-2%.). The air gap displacement is always in a range between 0 and 20% of the original size and for most of the time between 6% and 14%. In conclusion, one optimization model has been made and the simulation with all loads has been performed. All the changes bring an increase of mass equal to 30tons. With an increase of 18.7% of the original mass there is an improvement regarding the air gap behaviour equal to the 13.3%