Efficient Computation of Minmax Tests for Fault Isolation and Their Application to Structural Damage Localization

摘要: Fault detection and isolation can be handled by many different approaches. This paper builds upon a hypothesis test that checks whether the mean of Gaussian random vector has become non-zero in faulty state, based on chi2 test. For fault isolation, it to decided which components parameter set have changed, is done variants using so-called sensitivity minmax While only tested component taken into account approach, sensitivities all parameters are used leading better statistical properties at expense an increased computational burden. The computation respective variable cumbersome may ill-conditioned especially for large sets, asking hence careful numerical evaluation. Furthermore, procedure requires repetitive calculation each change, significant In this paper, dealing with problem, we propose new efficient variables, simultaneous QR decomposition parameters. Based scheme, set, decrease complexity one order magnitude total number Finally, show how useful structural damage localization, where asymptotically residual computed from output-only vibration data mechanical or civil structure.