Failure prediction and diagnosis control method applicable to shield tunneling machine

Abstract

The invention discloses a failure prediction and diagnosis control method applicable to a shield tunneling machine. The failure prediction and diagnosis control method comprises the following steps: (1) analyzing failure mechanism of the shield tunneling machine; (2) building a failure fault tree model of the shield tunneling machine and simultaneously determining failure logic relationships between all system members; (3) transforming a failure door into a discrete time Bayesian network according to the built failure fault tree model so as to obtain a shield tunneling machine failure discrete Bayesian network corresponding to the failure fault tree model; (4) performing failure prediction and risk diagnosis control on the shield tunneling machine by adopting a forward reasoning technology and a posterior probability of the Bayesian network.

Description

technical field [0001] The invention belongs to the related technical field of tunnel excavation equipment, and more specifically relates to a failure prediction and diagnosis control method applicable to a shield machine. Background technique [0002] In recent years, subway construction has developed rapidly in China. Due to the complex and changeable construction environment of the subway project, safety accidents occur frequently during the construction process, and the accidents caused by the failure of the shield machine account for almost half of them. In order to avoid serious casualties and property losses caused by construction safety accidents, engineering personnel at home and abroad have gradually paid attention to the failure risk management of shield machine excavation, and identified potential risks in the process of shield machine excavation based on qualitative or quantitative evaluation methods. At present, the fault tree analysis method is mainly used to...

AI Technical Summary

Problems solved by technology

However, the dynamic fault tree still has the following disadvantages: 1. The calculation workload is large, which can easily cause the problem of state combination explosion. This is because the solution of the existing dynamic fault tree is based on Markov (Markov) chain, and there is a possibility of combination explosion Defects, it is difficult to apply to large-scale sequential systems; 2. Under certain failure modes of the system, the dynamic fault tree cannot perform risk diagnosis and analysis, and it is difficult to identify the risk incentives that lead to risk events

Images