Concept of the simulation environment for prototyping design of fault detection and diagnostic systems

Abstract In recent years a number of new methods for fault detection and diagnosis (FDD) have been proposed in the literature. However, for a practitioner wishing to solve a concrete FDD problem it is often not easy to choose the right method for the problem at hand. Since the FDD design is normally iterative, use of CAD support is of great practical interest in order to make the design more efficient. In this paper we deal with the preliminary FDD design process referred to as prototyping in which simulation plays an important role. First, the main stages of the prototyping design cycle are identified. Accordingly, a concept is proposed for the simulation environment to support the needs of the designer during realisation of particular stages of the prototyping design cycle. Practical realisation of the concept is, however, confined by several problems. One of these is related to the fact that the FDD design often imposes processing of different knowledge items like heuristic facts, qualitative models or analytical models. To support their processing, it is necessary to make use of different program tools which, at the present level of development, lack the cost-effective mechanisms for easy integration. In turn, such a mismatch demands an extra effort from the designer in arranging transfer of data between the packages, as well as in handling the various tools. These problems are illustrated by an example of the environment developed for support of the FDD prototyping design of a DC motor.