Bearings Intelligent Fault Diagnosis by 1-D Adder Neural Networks

作     者：Jian Tang Chao Wei Quanchang Li Yinjun Wang Xiaoxi Ding Wenbin Huang 

作者机构：College of Mechanical EngineeringChongqing UniversityChongqing 400044China Department of Mechanical and Materials EngineeringQueen’s UniversityKingstonCanada 

出 版 物：《Journal of Dynamics, Monitoring and Diagnostics》 (动力学、监测与诊断学报（英文）)

年 卷 期：2022年第1卷第3期

页      面：160-168页

学科分类：0711[理学-系统科学] 07[理学] 08[工学] 081101[工学-控制理论与控制工程] 0811[工学-控制科学与工程] 071102[理学-系统分析与集成] 081103[工学-系统工程] 

基　　金：support provided by the China National Key Research and Development Program of China under Grant 2019YFB2004300 the National Natural Science Foundation of China under Grant 51975065 and 51805051 

主　　题：adder neural network convolutional neural network fault diagnosis intelligent bearings l1-norm distance 

摘      要：Integrated with sensors,processors,and radio frequency(RF)communication modules,intelligent bearing could achieve the autonomous perception and autonomous decision-making,guarantying the safety and reliability during their ***,because of the resource limitations of the end device,processors in the intelligent bearing are unable to carry the computational load of deep learning models like convolutional neural network(CNN),which involves a great amount of multiplicative *** minimize the computation cost of the conventional CNN,based on the idea of AdderNet,a 1-D adder neural network with a wide first-layer kernel(WAddNN)suitable for bearing fault diagnosis is proposed in this *** proposed method uses the l1-norm distance between filters and input features as the output response,thus making the whole network almost free of multiplicative *** whole model takes the original signal as the input,uses a wide kernel in the first adder layer to extract features and suppress the high frequency noise,and then uses two layers of small kernels for nonlinear *** experimental comparison with CNN models of the same structure,WAddNN is able to achieve a similar accuracy as CNN models with significantly reduced computational *** proposed model provides a new fault diagnosis method for intelligent bearings with limited resources.