Fault Diagnosis and Fault Tolerant Control with Application on a Wind Turbine Low Speed Shaft Encoder

In recent years, individual pitch control has been developed for wind turbines, with the purpose of reducing blade and tower loads. Such algorithms depend on reliable sensor information. The azimuth angle sensor, which positions the wind turbine rotor in its rotation, is quite important. This sensor has to be correct as blade pitch actions should be different at different azimuth angle as the wind speed varies within the rotor field due to different phenomena. A scheme detecting faults in this sensor has previously been designed for the application of a high end fault diagnosis and fault tolerant control of wind turbines using a benchmark model. In this paper, the fault diagnosis scheme is improved and integrated with a fault accommodation scheme which enables and disables the individual pitch algorithm based on the fault detection. In this way, the blade and tower loads are not increased due to individual pitch control algorithm operating with faulty azimuth angle inputs. The proposed approach is evaluated on a wind turbine benchmark model, which is based on the FAST aero-elastic code provided by NREL.