Fault Diagnosis of Advanced Wind Turbine Benchmark using Interval-based ARRs and Observers

Abstract In this paper, the problem of the fault diagnosis of an advanced wind turbine benchmark will be addressed using analytical redundant relations (ARRs) and observers that considers uncertainty in a bounded context, using the so-called interval approach. The fault detection test is based on checking the consistency between the measurements and the model by finding if the formers are inside the interval bounds provided by the interval model. In case a fault is detected, using the theoretical fault signature matrix against the full set of residuals available on-line, the fault is isolated. Two fault isolation schemes are compared. One based in the classical column matching and the other one using the row-reasoning inspired in the DX fault diagnosis approach. Finally, the proposed approach will be tested using the advanced wind turbine benchmark proposed in the literature.