Modeling Dynamical Causal Interactions with Fuzzy Temporal Networks for Process Operation Support Systems

Abstract

Fossil Power Plants are faced with ever-increasing requirements for better quality, higher production profits, safer operation and stringent environment regulation. New technologies are required to reduce the operator’s cognitive load and to achieve more consistent operations. The research described in this work intended to develop an efficient reasoning methodology for operation support systems. The proposed approach is based on a novel fuzzy reasoning to deal uncertainty and time, know as Fuzzy Temporal Network (FTN). A FTN is a formal and systematic structure (DAG), used to model dynamical causal interactions between the occurrence of events. The mechanism of possibility propagation is based on Mamdani inference method (fuzzy logic control methodology). The proposed approach is applied to fossil power plant diagnosis through a case study: the diagnosis and prediction of events in the drum level system.