DFIG Wind Generators Fault Diagnosis Considering Parameter and Measurement Uncertainties

Accurate and fast fault detection and isolation/identification (FDI) in wind power generators minimizes the downtime of the units as further damages can be prevented. In this paper, a robust FDI method for doubly fed induction generator (DFIG) based wind generators based on bond-graph (BG) theory is proposed. The comprehensive model of BG model of DFIG-based wind generators is developed that includes electrical, mechanical, and hydraulic subsystems. To represent the parameter uncertainties in the BG model, linear fractional transformation is utilized, in which the uncertain part of the BG model is separated from the nominal part of the model. Based on the nominal part of the BG model, analytical redundancy relationships are derived that are used for FDI. The uncertain part of the BG model is also used to derive the thresholds that are used for distinguishing faulty cases from nonfaulty cases in the presence of uncertainty. Results obtained from the case study, demonstrate that the proposed method is capable of detecting and pinpointing faults in different components, quickly and accurately.