A fault diagnosis method for electronic products based on discretized multivalued extended d-matrix

Abstract

The invention provides an electronic product fault diagnosis method based on a discretized multi-value extended D matrix, and belongs to the technical field of electronic product fault diagnosis and testing. Including: obtaining the sample data of the electronic product to be diagnosed in normal state and each fault state; clustering the sample data obtained at each test point, obtaining the number of clusters and cluster centers under each test point, and establishing discretization At the same time, construct a discretized test vector; calculate the Manhattan distance with each row in the discretized multi-valued D matrix, find the row of the D matrix with the shortest Manhattan distance to the test vector, and the state corresponding to the row is The state of the electronic product diagnosed according to the discretized test vector. The invention uses K-means clustering to discretize the data of a single test point, constructs a multi-valued D matrix, and realizes fault diagnosis by calculating the Manhattan distance between the test vector and the multi-valued D matrix, thereby improving the application feasibility of the method.

Description

technical field [0001] The invention relates to an electronic product fault diagnosis method based on a discretized multi-value extended D matrix, and belongs to the technical field of electronic product fault diagnosis and testing. Background technique [0002] With the development of electronic product integration and functional diversification, the transmission relationship of fault signals is becoming more and more complicated, which increases the difficulty of product fault diagnosis. At present, in the fault diagnosis test of electronic products, the correlation matrix between the fault and the test is used, also known as the D matrix, and the signal transmission relationship between the fault and the test is represented by 0 / 1 value, 0 means no fault, 1 means there is a fault , the fault diagnosis method based on D matrix has the characteristics of simple calculation and high operation efficiency, and has been widely used in aviation, aerospace, shipbuilding, weaponry...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problem that the current electronic product fault diagnosis simply uses 0 / 1 binary value to represent the test data, but the important information in the test data cannot be fully utilized, the present invention proposes a kind of electronic product fault diagnosis based on the discretized multi-valued extended D matrix. Diagnosis method, the present invention uses K-means clustering to discretize the data of a single test point, constructs a multi-valued D matrix, and realizes fault diagnosis by calculating the Manhattan distance between the test vector and the multi-valued D matrix

Images