An Approach to Gear Fault Diagnostic Using LMD Multi-Scale Permutation Entropy Assisted ACRO-RBF Network
Le Duc Hieu
Faculty of Automobile Technology, Hanoi University of Industry, Vietnam.
This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
The new gear diagnosis approach based on local mean decomposition (LMD), multi-scale permutation entropy (MPE), and radial basis function network (RBFN) is suggested. The energy of gear vibration signal is going to change in different frequency bands when the fault of gear appears. The LMD method is used to preprocess different types of gear vibration signals, which have been decomposed into a limited number of product functions (PFs). Afterward, the MPE is employed in extracting characteristic information from gear fault signals. After extracting feature vectors by MPE, the RBFN is applied to automate the gear fault diagnostic. Moreover, to decrease the cost time and increase the accurateness of the RBFN. In this paper, the parameters of RBFN has been optimized by artificial chemical reaction optimization (ACRO) algorithm. When the training process was over, the finest parameters of ACRO-RBFN have been found, the result of the classification shows a higher accurateness and lower cost time compared with the genetic algorithm-radial basis function network (GA-RBFN), particle swarm optimization-radial basis function network (PSO-RBFN) and RBFN methods.