A new approach for fault detection in digital relays-based power system using Petri nets

Fault detection in power systems, from the viewpoint of its required speed and accuracy, needs to be investigated yet. The hidden faults resulted from malfunctioning of circuit breakers and incorrect warnings of relays and happening of several faults, are the main difficulties in monitoring of power system. In this paper, Petri nets have been used for modeling and location detection of faults in power systems. In this deductive method, using the information of protection system's situation, the estimation of the faulted sections has been modeled using Petri nets. This deductive process can be represented graphically based on Petri nets and executed using matrix operations. The inputs of fault diagnosis models are the acquired data by the Remote Terminal Units (RTU) of the Supervisory Control and Data Acquisition (SCADA) system including the relay's trip signals and the signal of circuit breaker's situation. Logical operand information of digital protection relays such as “pickup” and “operation” of protection devices are more reliable than that of SCADA in reflection of relays trip condition. They can be added to the information of SCADA based fault diagnosis models. By using Petri nets, the processing time of information is reduced and the precise of fault detection procedure is increased. Also the proposed approach provides hierarchical monitoring of power systems.