Use of Fuzzy Cause-Effect Digraph for Resolution Fault Diagnosis for Process Plants. 1. Fuzzy Cause-Effect Digraph

Abstract

In order to remain efficiently functioning, chemical factories make heavy use of automated systems, such as warning systems and instrumentations, to monitor process variables and to control deviations within an allowable range in production processes. A process abnormality occurs when process variables (such as temperature/pressure) or process parameters (such as catalyst activity) deviate from the designed allowable ranges. A new model graph called fuzzy cause-effect digraph (FCDG) is proposed. This model expresses quantitative deviations of variables from the normal values with fuzzy set. It uses dynamic constraints (confluences) which are converted to dynamic fuzzy relations to express the dynamic gain between the variables in a chemical process. This replaces the steady-state gain between the variables originally expressed with a +, {minus}, or 0 by signed directed graph (SDG). Using this FCDG model would eliminate spurious interpretations attributed to system compensations and inverse responses from backward loops and forward paths in the process. The basic idea and development of this proposed methods are described in this paper. Moreover, this method can apply fuzzy reasoning to estimate the states of the unmeasured variables, to explain fault propagation paths, and to ascertain fault origins. The algorithm of fault diagnosis and its application proposedmore » in this paper are described in part 2.« less