A System Fault Diagnosis Method Based on Labeled Time Petri Net With Data

Regarding the fault diagnosis issue in real-time discrete event systems, existing detection methods typically involve modeling using labeled time Petri net (LTPN), followed by using the generated modified state class graph (MSCG) from LTPN to detect faults in the system. The diagnosis results are classified into normal, uncertain, and faulty. However, existing fault detection methods are based on time sequences for diagnosis; they ignore the changes in data flow in the system, making it impossible to determine whether there is a fault in the system when the detection result is uncertain. To address this issue, this letter proposes a modeling method of LTPN with data (LTPND), which binds corresponding data element operations on the transitions of LTPN to simulate the data flow in the system. Subsequently, the MSCG of LTPND is constructed, and a corresponding fault diagnosis algorithm is proposed based on MSCG. By traversing all the paths that satisfy the fire time and sequence information of observable transitions, faults in the system are detected from the perspectives of time constraints and data element changes, thus accurately judging the specific situations of uncertain states. Finally, the feasibility and effectiveness of the proposed method are validated through case analysis.