摘要
为提高轴承故障分类收敛速度和分类精度,提出一种动态调节学习率的堆叠自编码网络(SAE)。初始时刻给予一个较大的学习率,迭代过程中利用当前重构误差动态调节学习率的大小,根据重构误差梯度的正负值给出两种不同的学习率减小策略,使学习率大小更符合网络当前的运行状态,最后通过不同的有标签数据量进行反向微调,验证故障分类识别的准确率。实验结果表明:相比固定学习率,该动态调节学习率SAE网络预训练收敛时间减少17.70%,重构误差下降22.92%,故障分类准确率得到提高,且能在保持分类准确率的前提下,减少有标签样本量。
In order to obtain better convergence speed and classification accuracy of bearing fault classification,a Stacked Auto-Encoder(SAE)with dynamic adjustment of learning rate is proposed.In the following iteration,the gradient of the current reconstruction error is used to dynamically adjust the learning rate.According to the positive and negative value of the reconstruction error gradient,two different learning rate reduction strategies are given to make the learning rate more consistent with the current operation of the model.Finally,the accuracy of fault classification and recognition is verified by reverse fine-tuning of different labeled data quantities.The experimental results show that:compared with the fixed learning rate,the dynamic adjusted learning rate reduces 17.70%of the pre training convergence time,22.92%of the reconstruction error,improves the accuracy of fault classification,and reduces the number of labeled samples on the premise of maintaining the accuracy of classification.
作者
唐魏
郑源
潘虹
徐晶珺
TANG Wei;ZHENG Yuan;PAN Hong;XU Jingjun(College of Water Conservancy and Hydropower Engineering,Hohai University,Nanjing 210098,China;Institute of Innovation,Hohai University,Nanjing 210098,China;College of Energy and Electrical,Hohai University,Nanjing 211100,China)
出处
《计算机工程与应用》
CSCD
北大核心
2020年第20期264-269,共6页
Computer Engineering and Applications
基金
国家自然科学基金(No.51809082,No.51769035)。
关键词
自编码
深度学习
故障诊断
滚动轴承
动态学习率
auto-encoder
deep learning
fault diagnosis
rolling bearing
dynamic learning rate
