Equipment state prediction method based on fault tree information

Abstract

The invention discloses an equipment state prediction method based on fault tree information, comprising the following steps: firstly, searching a fault tree model, representing each event in the model by different variables, and representing a logic gate in the model by the structural relationship set of the variables; successively calculating fault reason variable prior probability distributionsets in a fault prediction model; calculating the conditional probability distribution of a fault mode variable and the conditional probability distribution of a fault transferring variable in the fault prediction model; representing the practical fault detection phenomenon of the equipment by a corresponding fault detection variable in the fault prediction model; according to incidence relation represented in the structural relationship set of the fault detection variables, calculating the conditional probability distribution of the fault detection variable in the fault prediction model; andaccording to the fault detection phenomenon real-time information collected on site, calculating equipment fault mode posterior probability state distribution. The invention can quickly and accurately predict the real-time state of the equipment, guides the monitoring and maintenance for equipment, effectively improves the maintenance efficiency and lowers the maintenance cost.

Description

technical field [0001] The invention belongs to the field of equipment maintenance and guarantee, and in particular relates to a method for predicting the state of equipment. Background technique [0002] With the rapid development of equipment design technology, equipment manufacturing technology and information technology, traditional equipment is gradually becoming integrated and intelligent, and its internal structure and correlation are becoming increasingly large, resulting in the status distribution of its components and the relationship between components of the equipment system. presents great complexity. It is difficult for maintenance personnel to predict the true state of the equipment, and under-maintenance or over-maintenance often occurs, which brings huge challenges to equipment maintenance and guarantee work. Equipment failure prediction is a multidisciplinary comprehensive technology involving machinery, electronics, materials, control, communication, comp...

AI Technical Summary

Problems solved by technology

[0015] In order to overcome the deficiency that the prior art cannot effectively predict faults, the present invention provides an equipment state prediction method based on fault tree information, which can quickly and accurately calculate the real-time operating state and reliability of the equipment, and is used to guide the maintenance plan of the equipment. Develop a stockpile of repaired spare parts

Images