Authors: Eid M. Al-Hajri and Dr. J.A. Rossiter
Continuous and discrete activities interact with each other in a hybrid system. This paper employs particle swarm algorithms and adaptive fuzzy systems to model and control the former that is represented with few places inside the Petri net model of such systems. It also addresses multi fault detection and isolation problem that is overcome using a new distributed interpreted Petri net based diagnoser. This diagnoser comprises a set of distributed diagnosers that are linked with multi sessions of the process to be monitored. The idea is randomly decomposing the incidence matrix of the central diagnoser into a
set of matrices with only one condition that no sub local incidence matrix has repeatable values. Unlike, central and distributed diagnosers, the proposed diagnoser is event detectable and has the ability to detect and isolate multiple faults. A real industrial oil producing station is used for testing the soundness of the proposed schemes.