Active Fault Tolerant Control of Wind Turbine Systems using Disturbance Observer-based Sliding Mode and Time Delay Estimation

In this paper, an active fault tolerant control based on time delay control, sliding mode, and nonlinear disturbance observer is proposed to control the pitch subsystem in the presence of actuator faults and uncertainties. Time delay estimation is applied as a fault estimation algorithm for detection and compensation. Then, a robust control law is synthesized to nullify uncertainty and fault effects using a combination of sliding mode, disturbance observer, and time delay with novel adaptation laws. In order to mitigate chattering which comes from the discontinuous control term, a nonlinear disturbance observer is designed. Through the proposed structure, the discontinuous gain is reduced significantly which leads to chattering reduction. Stability analysis is conducted through Lyapunov Theory. Moreover, wind speed profiles are generated using TurbSim, and simulations are performed based on a nonlinear two-mass wind turbine model and implemented in the FAST environment to verify the validity of the designed controllers. Finally, results reveal the effectiveness of the proposed controller compared to feedback linearization and gain-schedule proportional-integral controllers in the presence of uncertainty and different actuator faults such as hydraulic leakage, pump wear, and high air content in the oil.