A new approach to power system fault diagnosis based on fuzzy temporal order Petri nets

Abstract

After a power system failure, a large number of alarms are generated and uploaded to the dispatching center. Based on this information, a fast and accurate fault diagnosis method can provide key decision support for further failure emergency treatment. Existing fault diagnosis methods considering timing information are complicated and temporal reasoning is time-consuming. A new approach to power system fault diagnosis based on fuzzy temporal order Petri net is proposed. The temporal order of alarms information is directly combined with the hierarchical Petri net model. Temporal constraints of operations of protective relays and circuit breakers are also exploited. An automatic model construction algorithm is presented to improve the applicability of the proposed method. The reasoning efficiency is enhanced by introducing the layered matrix reasoning algorithm. After the fault diagnosis result is obtained, a backward reasoning process can be conducted to evaluate alarms information. Finally, the proposed method is tested on multiple cases and simulation results show that the all faults are correctly identified even when information distortion exists.

Keywords

• Petri net
• Fault diagnosis
• Network topology
• Fuzzy reasoning
• Power system
• Temporal reasoning