COMPOSING MODEL-BASED DIAGNOSTIC SYSTEMS

We present a novel approach to developing model-based diagnostic syste ms based on composing a solution from a prespecified and predeveloped, and hence reusable, set of software modules identified from a functio nal analysis of the complete model-based diagnostic process. This appr oach leads to a generic architecture, developed within the ARTIST proj ect (Leitch et al., 1992), that encompasses the main model-based diagn ostic techniques currently being employed in applications. The archite cture consists of predictor, candidate proposer, and diagnostic strate gist modules that represent the primitive functionalities of: applicat ion systems. Instantiations of each of these functionalities are provi ded for the main strategies, i.e., dependency recording and iterative search based techniques. From these modules three complete diagnostic systems have been developed and applied to two full-scale industrial a pplications and a laboratory-scale process-rig. Such an approach allow s the reuse of existing modules, customized by the relevant domain kno wledge, resulting in a much reduced development time and making the po wer and generality of model-based applications to diagnosis technicall y and economically viable for a wide range of industrial applications. (C) 1995 John Wiley & Sons, Inc.