Petri net based method for positioning hidden danger in train door system

Abstract

The embodiment of the present invention provides a petri net based method for positioning hidden danger in atrain door system. The method mainly comprises: building a extended time possibility constraint petri net model for the train door system; and on the basis of an inversion process of the extended time possibility constraint petri net, building a state transition condition of the extended time possibility constraint petri net model, and on the basis of the state transition condition and by decoupling and precipitation treatment process, positioning the hidden danger in the train door system. According to the embodiment of the present invention, the method aims at a complex system with information omission or an observation data that possibly has an error, provides an abduction reasoning algorithm and precipitates a pure causal chain, and on the basis of the precipitated pure causal chain, the method further analyzes and detects a recessive state or event in an urban rail door system, wherein the state and event are possibly uncaptured due to an limitation of an observe condition, or are system hidden dangers of themselves.

Description

technical field [0001] The invention relates to the technical field of media communication, in particular to a method for locating hidden dangers of train door systems based on petri nets. Background technique [0002] With the vigorous development of rail transit in major cities, a large number of urban rail trains have been put into operation. As an important role in the operation of urban rail vehicles, the door system has a direct impact on the design of their form, switch mechanism, and their manufacturing and control. The safe operation status of urban rail vehicles and timely and accurate diagnosis of door failures are of great significance to ensure the safe and stable operation of rail transit systems. At present, scholars at home and abroad are increasingly using computational intelligence and knowledge engineering techniques and methods in fault diagnosis, such as expert systems, artificial neural networks, rough set theory, fuzzy set theory, Bayesian networks, ge...

AI Technical Summary

Problems solved by technology

For example, it is difficult to establish and maintain the knowledge rule base of the expert system; the artificial neural network needs a large number of representative samples for learning before it is used, and it does not have the ability to explain knowledge; when the rough set method deals with multiple faults, the decision table is very difficult. Huge, even the problem of combination explosion; the acquisition of membership degree and the construction method of complex system fuzzy model in fuzzy set theory are not perfect; the training of Bayesian network is more complicated and requires prior probability information, for more complex faults and The analysis of a large-scale system is actually an NP-complete problem; the genetic algorithm shows certain difficulties in establishing a reasonable fitness function and parameter selection

Images