Prognostic Method of Degradation Fault Based on Multi-fusion Artificial Intelligence
LI Yonglong
WANG Zhendong
ZHAI Yuting
Abstract：There are many problems in the prognosis of current equipment degradation faults,such as few historical monitor-ing data samples,difficult to collect degradation fault samples,static fault prediction and so on.Therefore,the prognostic method of degradation fault based on multi-fusion artificial intelligence is proposed,in which the dynamic update support vector regression can improve the accuracy of static data prediction,principal components analysis can reduce the dimensionality of multi-dimension-al data,and K-means clustering method can detect degradation faults without fault samples.The feasibility of the proposed method is verified by simulation experiments,in which the root mean square error of dynamic update support vector regression is 0.44947,K-means clustering can detect degradation fault samples on the basis of accurate data prediction.The experimental results show that the proposed method can effectively predict and alarm the equipment degradation faults.
Keywords：support vector regressionprincipal components analysisK-means clustering
Publication Date：2023-10-20
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：6( 61-66 )