Exceptional part locating method in multi-part engineering system

Abstract

The invention relates to an exceptional part locating method in a multi-part engineering system. The exceptional part locating method includes the step of building judgment standards, wherein the standards for judging whether sampling data of the system are exceptional or not are built based on historical sampling data; the step of collecting data in real time, wherein current sampling data of the system are acquired; the step of judging exceptions of the system, wherein whether the current sampling data of the system are exceptional or not is judged according to the standards for judging whether the sampling data of the system are exceptional or not; the step of calculating posterior probabilities of part exceptions, wherein if the current sampling data of the system are exceptional, the posterior probabilities of exceptions of parts in the system are calculated respectively according to a Bayes method; the step of determining the exceptional parts, wherein the exceptional parts in the system are determined according to the maximum posterior probability criterion. The locating method is simple in process and easy to achieve, the exceptional parts in the multi-part engineering system can be accurately located through a few number of sensors, and users can conveniently maintain the system.

Description

technical field [0001] The invention relates to the technical field of fault detection and diagnosis, in particular to a method for locating abnormal components in a multi-component engineering system. Background technique [0002] Abnormal component location is a powerful guarantee for the safe operation of complex engineering systems. Due to the limitations of the system space structure or operating costs, most of the existing complex engineering systems cannot fully monitor the various components of the entire system by deploying a large number of sensors, but only a small number of sensors can be deployed in the system to monitor The operating status of each component of the entire complex engineering system. [0003] Due to the limited number of sensors and the measurement errors contained in the sensors themselves, it is very difficult to locate abnormal components in the system. Abnormal component location technology is closely related to the type and quality of sen...

AI Technical Summary

Problems solved by technology

[0004] In the existing engineering system, in the face of the complex correspondence between sampling data and abnormal parts, as well as various uncertain factors in the process of locating abnormal parts, the positioning accuracy of traditional abnormal part positioning methods is low, and it is no longer possible to fulfil requirements

Images