Research On Fault Diagnosis Methods Of Sequential Discrete Event Mechatronic System Based On Petri Net

The function of automation equipment becomes more powerful, the structure becomes more complicated and the frequency of fault becomes higher with the quick development of modern science and technology, at the same time, there are a lot of discrete event mechatronic system with the character of instantaneousness, asynchronies, sequence, in many large and complicated mechatronic systems. The sequential discrete event mechatronic system is special, but it lacks of related research methods. In order to adapt to the requirement of reliability and stability, it is necessary to research the fault diagnosis methods of the sequential discrete event mechatronic system.The typical styles and characters of sequential discrete event mechatronic system were studied in this article, including serial sequence, select sequence and parallel sequence. The characteristics and principle of the fault of sequential discrete event mechatronic system and the related modeling were analyzed. Analyze the foundation of the Petri net modeling the system and establish the basic Petri net models for the three typical types based on the action characters of sequential discrete event mechatronic system. At the same time, consider the time as statistical information pulled in Petri net, and establish timed Petri net modeling for the sequential discrete event mechatronic system. The fault diagnosis arithmetic based on Bayes test method and the significance-test of time statistics of the transition was put forward. Under the guidance of modular fault diagnosis idea, stratify and locate the fault source combined with prior probability and posterior probability of the system. Analyzing the two types fault in this arithmetic, discussed and deduced the threshold probability when the misdiagnosis probability was least. The work characteristics of micro-drill's detection equipment were studied in this paper, and the feasibility of considering it as the test platform of the sequential discrete event mechatronic system was analyzed. Fault forecast methods based on fuzzy reasoning were studied and put forward, and establish the fuzzy Petri net model for the sequential discrete event mechatronic system, and deduce the fuzzy membership functions calculation method of tokens of the places, based on fault diagnosis.The test result of micro-drill's detection equipment shows that the fault diagnosis and forecast methods are feasible, and the accuracy rate of fault diagnosis is high, and the research result in this dissertation provides technical reference for the other industrial fault diagnosis with the characters of discretization.