Open AccessFault Prediction of Electromagnetic Brake Based on AINN and Grey MGM(1,n) Model
Author(s) -
Shoujun Li,
Guangyu Li,
Zhiming Wang
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1626/1/012010
Subject(s) - fault (geology) , artificial neural network , brake , cluster analysis , convergence (economics) , fuzzy logic , computer science , feature (linguistics) , identification (biology) , pattern recognition (psychology) , engineering , data mining , algorithm , artificial intelligence , automotive engineering , linguistics , philosophy , seismology , economic growth , economics , geology , botany , biology
In order to predict the possible fault points of electromagnetic brake, a novel combined diagnosis model based on multivariable grey prediction model MGM(1, n ) and adaptive integrated neural network (AINN) was proposed in this paper. MGM(1, n ) model was used to predict the development trend of multiple groups of fault feature factors. Then by Fuzzy C-mean (FCM) algorithm, cluster analysis on the predicted results from MGM(1,1) was performed, and in turn, clustering results feed into of AINN to predict and identify fault features. To verify the performance of the combined model, a case of electronic brake fault diagnosis is given, and the model analysis consisting of 8 possible faults is carried out, finding that the proposed combined model has better fault identification capacity with better convergence and less error, showing stronger promotional practical value.