Fault detection method based on ECA rule

Abstract

The invention relates to a high-speed train fault detection method based on an ECA rule. To begin with, the ECA rule is utilized to carry out modeling on faults of a network control system of a high-speed train, so that a complex sequential relationship between fault data can be described, and the fault description ability is enhanced; then, since the ECA rule is suitable for describing active behaviors of the system, an active working mode of " event detecting-condition judging-action triggering" of a fault diagnosis system can be supported, and the defect of a conventional "request-response " mode is overcome; meanwhile, heuristic search based on rule graphs is utilized in the method, so that the occurred faults can be diagnosed as many as possible in a given time based on collected data; and since the input of the fault diagnosis based on the ECA rule is atomic events but not original collected data, data irrelevant to the diagnosis rules is filtered out, and the processing efficiency of the fault diagnosis system under the condition of mass collected data is improved.

Description

technical field [0001] The patent of the present invention relates to fault detection technology, in particular to a fault detection method based on ECA rules of a high-speed train network control system. Background technique [0002] ECA (Event-Condition-Action) rules have strong semantic expression ability, and can support various applications such as integrity maintenance, derived data maintenance, production monitoring, market monitoring, and decision support systems. ECA is an active service mechanism for active database systems. has been widely accepted. [0003] With the rapid development of computer technology and automation control technology, the scale of various application systems has expanded, the difficulty of control has increased, and the correlation between various components has become more and more complex. The occurrence of a fault may have a huge impact on the system or even cause system failure. , so enterprises urgently need a sensitive and accurate f...

AI Technical Summary

Problems solved by technology

However, in fault diagnosis, it is first necessary to extract device feature information. In order to accurately describe faults in many scenarios, in addition to paying attention to the quantity and logical relationship of feature information, the timing relationship between information also needs to be carefully considered. The production formula The description of such relationships in the rules is complex and vague

[0010] (2) Efficiency reduction under large-scale data: In actual application scenarios, the amount of data is often quite huge, and there are often data irrelevant to diagnosis. Since traditional fault expert diagnosis systems do not filter these data, these redundancy The influx of data into the diagnostic system reduces the efficiency of fault diagnosis

[0011] (3) Lack of initiative: The traditional fault diagnosis system is a request-response mode, which makes it lack of initiative

Images