Wind turbine pitch system condition monitoring based on performance curves in multiple states

The pitch system playing a key role to capture wind energy is one of the most critical components of wind turbines (WTs). Meanwhile, it suffers the highest failure rate and major downtime. Improving the capacity of pitch system condition monitoring technique can effectively reduce operation and maintenance (O&M) costs. Supervisory control and data acquisition (SCADA) data including comprehensive signal information from most subassemblies have been widely applied to condition monitoring and fault diagnosis of WTs. However, because of the complex operational condition of WTs, SCADA data become more complicated to study. This paper proposes a condition monitoring method of pitch system in multiple states based on blade angle-rotor speed-wind speed curve. In order to make the performance curve easier to explain, SCADA data are divided into nine parts according to the operational condition of WTs. Gaussian mixture model clustering, mean Euclidean distance and line fitting are applied to detect anomalies in three main states. Four faults in three categories have been study to demonstrate the feasibility of proposed method. It has been shown that the proposed condition monitoring method can not only simplify performance curves, but also detect pitch system faults effectively.