《机电工程》杂志社科技期刊核心期刊论文投稿机电一体化机械、仪表电工自动化、计算机浙江大学主办

Introduction

International Standard Serial Number:

ISSN 1001-4551

Sponsor:

Zhejiang University;

Zhejiang Machinery and Electrical Group

Edited by:

Editorial of Journal of Mechanical & Electrical Engineering

Chief Editor:

ZHAO Qun

Vice Chief Editor:

TANG ren-zhong,

LUO Xiang-yang

Tel:

86-571-87041360,87239525

Fax:

86-571-87239571

Add:

No.9 Gaoguannong,Daxue Road,Hangzhou,China

P.C:

310009

E-mail:

meem_contribute@163.com

Remaining life prediction of rolling bearing based on CNN-BiLSTM model with attention mechanism

Published:2021-12-22 author:ZHAO Guang-qian, JIANG Pei-gang, LIN Tian-ran Browse: 1564 Check PDF documents

Remaining life prediction of rolling bearing based on

CNN-BiLSTM model with attention mechanism

ZHAO Guang-qian, JIANG Pei-gang, LIN Tian-ran

(School of Mechanical and Automotive Engineering, Qingdao University

of Technology, Qingdao 266520, China)

Abstract: In order to make full use of the temporal information of data to accurately predict the remaining useful life(RUL) of rolling element bearing, an intelligent rolling element bearing RUL prediction method based on convolution neural network (CNN) and bidirectional long short-term memory network (BiLSTM) was proposed. Firstly, 12 timedomain features and 4 frequency-domain features were extracted from the data as inputs to the neural network. Then, a CNN-BiLSTM algorithm based on attention mechanism (AM) was proposed to extract the degradation chracteristics from the input data and further solve the information loss problem of BiLSTM. Finally, the validity of the proposed method was verified by PHM 2012 bearings degradation dataset published in IEEE PHM 2012 Data Challenge. The advantage of the proposed method is examined by comparing the prediction result with those using other popular RUL prediction techniques such as FCNN, CNN-BiLSTM and CNN-LSTM-AM algorithms. The research results show that comparing with other methods, the RMSE values of rolling element bearings RUL prediction of the proposed method can be reduced by 25.85%, 7.32% and 10.59% respectively, and the Score value can be increased by 3.65%, 2.12% and 1.58% respectively, which verifies the superiority of this method in the application of rolling element bearings RUL prediction.

Key words: rolling element bearing; remaining useful life (RUL); convolutional neural networks(CNN); bidirectional long short-term memory(BiLSTM); attention mechanism (AM)

ZHAO Guang-qian, JIANG Pei-gang, LIN Tian-ran. Remaining life prediction of rolling bearing based on CNNBiLSTM model with attention mechanism［J］.Journal of Mechanical & Electrical Engineering, 2021,38(10):1253-1260.

Advanced Search

Online Office System

Chinese Core Periodicals
Chinese Sci-tech Core Periodicals
SA, INSPEC Indexed
CSA: T Indexed
UPD:Indexed