Sequential testing algorithms for multiple fault diagnosis

A common simplifying assumption made by the existing testability analysis a nd diagnostic tools is that there exists, at most, a single fault in the sy stem at any given time. This assumption does not hold for complex systems w ith large numbers of components and/or systems with little or no opportunit y for maintenance during operation. In this paper, we consider the problem of constructing optimal and near-optimal test sequences for multiple fault diagnosis. The computational complexity of solving the optimal multiple-fau lt isolation problem is super exponential, that is, it is much more difficu lt than the single-fault isolation problem, which, by itself, is NP-hard(1) [1]. By employing concepts from information theory and AND/OR graph search and by exploiting the single fault testing strategies of [1], we present s everal test sequencing algorithms for the multiple fault isolation problem. These algorithms provide a tradeoff between the degree of suboptimality an d computational complexity. Furthermore, we present novel diagnostic strate gies that generate a diagnostic directed graph (digraph), instead of a trad itional diagnostic tree, for multiple fault diagnosis, Using this approach, the storage complexity of the overall diagnostic strategy reduces substant ially, The algorithms developed herein have been successfully applied to se veral real-world systems. Computational results indicate that the size of a multiple fault diagnostic strategy is strictly related to the structure of the system.