Simulation method of fault diagnosis tree evaluation

Autor: Junyou Shi, Kaiyue Lv
Rok vydání: 2012
Předmět:
Zdroj: Proceedings of the IEEE 2012 Prognostics and System Health Management Conference (PHM-2012 Beijing).
DOI: 10.1109/phm.2012.6228787
Popis: Fault diagnosis tree based on dependency models is becoming an important way to improve testability. For the purpose of evaluation of a fault diagnosis tree, a simulation method of fault diagnosis tree evaluation is proposed, with which we can evaluate the real diagnosis capacity of a fault diagnosis tree. The principal of fault diagnosis tree evaluation is analyzed, with the tuple relation models of simulation established and the detailed simulation flow given. Evaluations of fault detection rate (FDR), fault isolation rate (FIR) can be made with this method. The main steps of this method include: 1) configure test thresholds, 2) obtain original test data, 3) convert original test data into logical test data, 4) simulate fault diagnosis tree and obtain diagnostic conclusions, 5) evaluate diagnosis capacity and give corresponding suggestions. Finally, a case of the signal conditioning circuit is given. Based on the method mentioned above, assistant software is developed and the simulation is made with test data in fault states, and the result shows that the method is feasible and effective.
Databáze: OpenAIRE