A feature selection method, GRRF-A, based on guided regularized random forest is proposed in this paper, aiming at solving marine diesel engine fault problem. GRRF-A improves GRRF algorithm according to the characteristics of marine diesel engine by using mean decrease in accuracy as importance measure, which is proven to be effective. The computation cost of GRRF-A is small and its results are relatively accurate and stable, which makes this method more suitable for marine diesel engine diagnosis than other methods discussed in this paper.
Research on fault pattern analysis of marine diesel engine based on random forest algorithm
01.08.2017
364554 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Research on Fault Diagnosis of Marine Diesel Engine Based on KFDA
Tema Archiv | 2012
|Research on fault prediction of marine diesel engine based on attention-LSTM
SAGE Publications | 2023
|Large marine diesel engine research
Engineering Index Backfile | 1940
|Real-time marine diesel engine simulation for fault diagnosis
Tema Archiv | 2002
|Real-Time Marine Diesel Engine Simulation for Fault Diagnosis
Online Contents | 2002
|