Tuning and comparing fault diagnosis methods for aeronautical systems via kriging-based optimization

Abstract

Many approaches address fault detection and isolation (FDI) based on analytical redundancy. To rank them, it is necessary to de¦ne performance indices and realistic sets of test cases on which they will be evaluated. For the ranking to be fair, each of the methods under consideration should have its internal parameters tuned optimally. The work presented uses a combination of tools developed in the context of computer experiments to achieve this tuning from a limited number of numerical evaluations. The methodology is then extended so as to provide a robust tuning in the worst-case sense.