基于改进ResNet-18模型的弧齿锥齿轮安装位置偏差分类识别方法研究  

Research on classification and identification method of installation position deviation of spiral bevel gears based on improved ResNet-18 model

作　　者：刘小凯 边骥轩 张江勇 董岑鑫 王卫军 LIU Xiaokai;BIAN Jixuan;ZHANG Jiangyong;DONG Cenxin;WANG Weijun(Intelligent Robot&Equipment Center,Guangzhou Institute of Advanced Technology,Guangzhou 511458,Guangdong,China;China Northern Vehicle Research Institute,Beijing 100072,China)

机构地区：[1]广州先进技术研究所机器人与智能装备中心,广东广州511458 [2]中国北方车辆研究所,北京100072

出　　处：《农业装备与车辆工程》2024年第3期96-100,共5页Agricultural Equipment & Vehicle Engineering

基　　金：国家重点研发计划项目子课题(2018YFA0902901)。

摘　　要：针对现有的弧齿锥齿轮安装位置偏差分类识别效率低的问题,使用改进的ResNet-18神经网络构建轻量级弧齿锥齿轮安装位置偏差分类识别模型,改进的网络修改了ResNet-18网络第一卷积层,并引入了通道与空间注意力机制。实验结果表明,改进ResNet-18较原网络对于齿面接触印痕图像的分类准确率提高了3.23%,损失值降低了0.04。该方法为齿轮安装偏差识别提供了新路径,对弧齿锥齿轮总成的安装调整作业提供了一定的指导。Aiming at the problems of low efficiency in identification of spiral bevel gear installation position deviation,the improved ResNet-18 neural network was used to build a light-weight spiral bevel gear installation position deviation classification model.The improved network modified the first convolutional layer of the ResNet-18 network,and introduced a channel and spatial attention mechanism.The experimental results showed that,for gear contact pattern identification,the accuracy of the improved ResNet-18 was 3.23%higher than that of the original network,and the loss value was reduced by 0.04.The proposed method offers a new path for gear installation deviation identification,and provides guidance in assembling the spiral bevel gear.

关 键 词：弧齿锥齿轮 接触印痕 安装位置偏差 注意力机制 ResNet-18 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TH132.41[自动化与计算机技术—计算机科学与技术]