Fuzzy reasoning in co-operative supervision systems

This paper considers a decision support system dedicated to fault detection and isolation from a human-machine co-operation point of view. Detection a nd isolation are based on different models of the process (non-linear and l inear causal local models). Reasoning using real numbers is often used by h uman beings; fuzzy logic is introduced as a numerical-symbolic interface be tween the quantitative fault indicators and the symbolic diagnostic reasoni ng on them; it also provides an effective decision-making tool in imprecise or uncertain environments while managing model uncertainty, sensor impreci sion and vague normal behavior limits. Fuzzy rules are modelled geometrical ly; fuzzy sets are represented as points in a description space. A prototyp e graphical interface with structural, causal and historical views gives co mplete information to the human operator. In such an interface, fuzziness i s displayed as a colour palette evolving with time. (C) 2000 Elsevier Scien ce Ltd. All rights reserved.