Decoupling Adaptive Sliding Mode Observer Design for Wind Turbines Subject to Simultaneous Faults in Sensors and Actuators

• Published in: IEEE/CAA journal of automatica sinica Vol. 8; no. 4; pp. 837 - 847
• Main Authors: Habibi, Hamed, Howard, Ian, Simani, Silvio, Fekih, Afef
• Format: Journal Article
• Language: English
• Published: Piscataway The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 01.04.2021; School of Civil and Mechanical Engineering, Curtin University, WA 6102, Australia%Department of Engineering, University of Ferrara, Ferrara 44100, Italy%Department of Electrical and Computer Engineering, University of Louisiana at Lafayette, Louisiana 70504 USA
• Subjects: Actuators; Decoupling; Disturbances; Estimation; Fault detection; Fault tolerance; Fault tolerant control; Fault tolerant systems; horizontal axis wind turbines; Monte Carlo simulation; Monte-Carlo analysis; Observers; principle of separation; Reliability analysis; Sensors; simultaneous faults; Sliding mode control; sliding mode observer; Wind speed; Wind turbines; horizontal axis wind turbines; simultaneous faults; sliding mode observer; principle of separation; Fault tolerant control; Monte-Carlo analysis
• ISSN: 2329-9266; 2329-9274
• DOI: 10.1109/JAS.2021.1003931

This paper proposes an adaptive sliding mode observer (ASMO)-based approach for wind turbines subject to simultaneous faults in sensors and actuators. The proposed approach enables the simultaneous detection of actuator and sensor faults without the need for any redundant hardware components. Additionally, wind speed variations are considered as unknown disturbances, thus eliminating the need for accurate measurement or estimation. The proposed ASMO enables the accurate estimation and reconstruction of the descriptor states and disturbances. The proposed design implements the principle of separation to enable the use of the nominal controller during faulty conditions. Fault tolerance is achieved by implementing a signal correction scheme to recover the nominal behavior. The performance of the proposed approach is validated using a 4.8 MW wind turbine benchmark model subject to various faults. Monte-Carlo analysis is also carried out to further evaluate the reliability and robustness of the proposed approach in the presence of measurement errors. Simplicity, ease of implementation and the decoupling property are among the positive features of the proposed approach.