基于双边矢量概率矩阵的故障诊断方法研究

Research on a Fault Diagnosis Method Based OnBilateral Vector Probability Matrix

摘要:

针对复杂装备中存在的错综复杂、关联耦合的相互关系以及大量的不确定因素和不确定信息,且测试样本数据呈现“小子样”、“不完备”的特点,研究以FMEA报告为基础,获取双边相关性矩阵,利用测试项目和故障模式的相关性强弱和测试顺序构建矢量矩阵,并通过模糊层次分析法和贝叶斯理论获取双边矢量矩阵概率参数,解决测试数据“不确定”、“小子样”、“不完备”情况下的故障定位问题。

Abstract:

Aiming at the intricate and correlated interrelationships in complex equipment, as well as a large number of uncertain factors and uncertain information, and the characteristics of the test sample data are "small sample" and "incomplete", this article uses the FMEA report as the basis to obtain bilateral information. Correlation matrix, the vector matrix is constructed by using the correlation between test items and failure modes and the test sequence, and the probability parameters of the bilateral vector matrix are obtained by fuzzy analytic hierarchy process and Bayesian theory to solve the test data “uncertain”, “Small sample”, "Incomplete" fault location problem.