Software for fault diagnosis using knowledge models in petri nets

Fault diagnosis systems in electric sector companies require precision and flexibility properties in case of events. Currently, there are systems aimed at improving the diagnosis process through various methods and techniques, reducing response time to disturbances. However, few proposals unify graphical models of knowledge with process signals. These signals can be provided by devices such as programmable logic controllers (PLCs). This article proposes novel model-driven software based on Petri nets and integrated with process signals for fault diagnosis in power plants. A case study demonstrates the flexibility and adaptability of the software when new concepts change in the knowledge models, without requiring reengineering procedures to be performed on the software.