Information criteria for residual generation and fault detection and isolation (Q1360454)

In many applications the problem of fault detection and isolation (FDI) is a crucial issue that has been theoretically and experimentally investigated with different types of approaches. Using an information point of view, the author discusses deterministic versus stochastic tools for residual generation and evaluation for FDI in linear time-invariant (LTI) state-space systems. It has been widely acknowledged that the FDI problem can be split into two steps: generation of residuals, and residual evaluation, namely design of decision rules, based on these residuals. In both types of approaches to off-line FDI, residual generation can be viewed as the design of a linear transformation of a Gaussian vector. Several statistical isolation methods are revisited, using both a linear-transform formulation and the information content of the corresponding residuals. Several multiple-fault cases, with or without causality assumptions are considered, and an optimality criterion for the most general one is discussed. New information criteria are proposed for investigated the residual optimization problem. The last section of this important paper is devoted to conclusions and questions open for future research.