Fault diagnosis in wind turbines based on ANFIS and Takagi–Sugeno interval observers

作者：

Highlights：

• A wind turbine uncertain model is obtained based on sensor data and an ANFIS method.

• Based on the ANFIS model, an uncertain Takagi–Sugeno (TS) model is obtained.

• A set of interval TS observers is designed to detect and isolate sensor faults.

• The method is based on healthy data and does not require a mathematical model.

• Numerical experiments are performed in a well-accepted wind turbine benchmark.

摘要

•A wind turbine uncertain model is obtained based on sensor data and an ANFIS method.•Based on the ANFIS model, an uncertain Takagi–Sugeno (TS) model is obtained.•A set of interval TS observers is designed to detect and isolate sensor faults.•The method is based on healthy data and does not require a mathematical model.•Numerical experiments are performed in a well-accepted wind turbine benchmark.

论文关键词：Wind turbine control,Fault detection,Adaptive Neuro-Fuzzy Inference System,Takagi–Sugeno models,Interval observers

论文评审过程：Received 15 March 2022, Revised 11 May 2022, Accepted 28 May 2022, Available online 9 June 2022, Version of Record 21 June 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.117698