Dynamic causal model diagnostic reasoning for online technical process supervision

Model-based diagnosis is founded on the construction of fault indicators. T he methods proposed for this purpose generally represent the process by mea ns of an extremely inflexible formalism that limits the scope of applicatio ns. Moreover, it is usually difficult and costly to develop precise mathema tical models of complex plants. New and more flexible techniques intended n otably to explain the observed behavior open new perspectives for fault det ection and diagnosis. The diagnostic procedures for such plants are general ly integrated into a supervisory system, and must therefore be provided wit h explanatory features that are essential interpretation and decision-makin g supports. Techniques based on causal graphs constitute a promising approa ch for this purpose. A causal graph represents the process at a high level of abstraction, and may be adapted to a variety of modeling knowledge corre sponding to different degrees of precision in the underlying mathematical m odels. When the process is dynamic the causal structure must allow temporal reasoning. Lastly, because reasoning on real numbers is often used by huma n beings, fuzzy logic is introduced as a numeric-symbolic interface between the quantitative fault indicators and the symbolic diagnostic reasoning on them; it also provides an effective decision-making tool in imprecise or u ncertain environments. An industrial application in the nuclear fuel reproc essing industry is presented. (C) 2000 Elsevier Science Ltd. All rights res erved.