Repairable GO algorithm based on dynamic Bayesian network

Abstract

A repairable GO algorithm based on a dynamic Bayesian network comprises the steps of: I, modeling a system into a dynamic repairable GO method model diagram comprising a repairable operator, II, mapping a dynamic repairable part as a dynamic Bayesian network node, and determining a conditional probability table of the dynamic Bayesian network node, and III, determining a set membership of the dynamic Bayesian network node, making calculation through the mature Bayesian network, and finally obtaining a relevant result of reliability modeling, and, if the repairable GO method comprises a static operator, further comprises the steps of: IV, mapping an unrepairable part as a Bayesian network node and determining a conditional probability table of the Bayesian network node, V, determining a node after conversion and a set membership of a Bayesian network node corresponding to an input signal of the node, VI, combining a dynamic Bayesian node and a static node into the complete Bayesian network, and VII, making calculation through the mature Bayesian network to obtain a relevant result of the reliability modeling.

Description

technical field [0001] The invention provides a modifiable GO method algorithm based on a dynamic Bayesian network, which can be applied to multi-state and time-sequential systems in industrial systems. For systems with strong timing in engineering, the GO method is generally selected for modeling. The new algorithm can realize fast calculation and analysis of the GO method through simple logic conversion rules. The technical field of the present invention is the field of computer-aided reliability design and analysis. Background technique [0002] With the expansion of the complexity and scale of weaponry, it is often necessary to assign and verify the relevant reliability parameters of the product through modeling before it is put into production. Moreover, in practical engineering applications, the equipment is often not disposable, and it is often necessary to repair the equipment after the equipment fails, which puts forward high requirements for the GO method modeling...

AI Technical Summary

Problems solved by technology

[0004] The present invention is based on a dynamic Bayesian network-based modifiable GO method algorithm. Its purpose and problem are to solve the problem that the existing GO method modeling and analysis software cannot handle common signals, and cannot efficiently process the analysis of complex models. and calculations, problems with correct calculations for repairable parts

Images