Structural similarity matching method for fault tree

Abstract

The invention discloses a structural similarity matching method for a fault tree. The structural similarity matching method comprises the steps of: calculating an extended cut set of a case fault tree, calculating an extended cut set of a standard fault tree, matching the extended cut set of the case fault tree with the extended cut set of the standard fault tree, and judging which type of standard fault trees does the case fault tree belongs to; and regarding a set of portions of unmatched standard tree extended cut set as a structure absence event and recommending the structure absence event to an analyst, and complementing the integrity of a case tree structure. The structural similarity matching method can identify the case fault tree type of a specific accident, performs similarity matching on the structure in an accident case fault tree with the structure in the standard fault tree, conducts accident analysis more precisely and quickly, provides powerful support for accident analysis, prediction and early warning in the safety-critical field, is conducive to improving the early warning and prevention capability of enterprises for major accidents, reduces property losses, and prevents the damage caused by the accident.

Description

technical field [0001] The invention relates to the technical field of accident analysis, in particular to a fault tree-oriented structural similarity matching method. Background technique [0002] With the increasing degree of integration, automation, and scale in safety-critical fields such as high-speed railways, chemical production, and aerospace, the impact and harm of accident consequences are also becoming more and more serious. Accurate and comprehensive analysis of the accidents that have occurred and drawing experience from the accidents that have occurred are important ways to guide safe production, accident prevention and early warning. Therefore, accident analysts are required to carefully analyze each accident link, and analyze the accident as comprehensively and accurately as possible. However, due to the limitations of accident analysts’ own experience and the limitations of accident investigation, although there are a large number of expert experience in ac...

AI Technical Summary

Problems solved by technology

[0005] Existing accident analysis technologies are only based on safety analysis methods such as fault tree, HAZOP, FME(C)A and event tree to conduct cause analysis, accident evolution analysis, and consequence prediction of complex accident models. These technologies can provide a framework to guide experts in accident analysis. In-depth analysis can also be carried out on the results of expert analysis, but these technologies cannot provide technical support in terms of reusing expert experience, providing automatic identification of accident types, and judgment of the integrity of accident analysis. No research has been found on how to provide analysis technology to automatically use expert experience Accumulate case studies supporting and supplementing new incidents

Images