International Journal of Advances in Computer Science and Its Applications
Author(s) : HUIQIANG WANG, MICHAEL WEYRICH, NASSER JAZDI
At present most of fault diagnosis systems are dedicated to fault detection, fault elimination and fault effect analysis with specific diagnosis approaches. Even though fault effect based on these approaches concerns the affected components as well as functions. However, there is no analysis of the still available functions which could continue to operate despite the failure. Currently, there are few approaches working on the fault analysis based on the existing fault diagnosis system. To support a full fault diagnosis, this paper proposes a novel full fault effect analysis approach based on system models. Within a defect component in an automation system, available functions can be identified by the presented approach. The presented approach uses the results of the existing fault diagnosis system as an analysis basis, e.g. the fault ID and the fault location. The propagation of a fault is identified with the help of the requirement-function-component models, which are provided by the manufacturer. As a result, the presented approach is able to identify the available functions in case of the appearance of defect components. In this way, it enables an industrial automation system to continue its operation with restricted functionality despite a fault.