| 引用本文: | 魏文军,张轩铭,杨立本.基于模糊聚类和改进Densenet网络的小样本轴承故障诊断[J].哈尔滨工业大学学报,2024,56(3):154.DOI:10.11918/202206075 |
| WEI Wenjun,ZHANG Xuanming,YANG Liben.Fault diagnosis of small sample bearings based on fuzzy clustering and improved Densenet network[J].Journal of Harbin Institute of Technology,2024,56(3):154.DOI:10.11918/202206075 |
|
| 摘要: |
| 针对实际中轴承的故障数据少难以满足深度学习数据大量训练模型的要求,利用卷积神经网络的微小特征提取优势和模糊聚类不需要训练即可完成分类的特点,提出了一种基于模糊聚类和改进Densenet网络的小样本轴承故障诊断方法。首先将预训练微调的Densenet网络去掉分类只保留特征提取层,设计一个维度自适应全局均值池化层（GAP）代替全连接层（FC）,其次利用模糊聚类代替Densenet网络的softmax分类层,不需要训练即可完成分类。实验结果表明:该算法利用小样本数据训练网络中的GAP参数,模型需要的训练样本大大减少,诊断时将轴承时域图像输入到网络中,在GAP层输出1 920个特征数据,不同故障状态的特征数据构建特征向量矩阵,利用模糊聚类方法求得模糊相似矩阵和模糊等价矩阵,当置信因子从大到小变化时,由对应布尔矩阵得到动态聚类图,从而实现轴承故障分类。 |
| 关键词: 小样本 全局均值池化层 迁移学习 模糊聚类 故障诊断 |
| DOI：10.11918/202206075 |
| 分类号:TP391;U298 |
| 文献标识码:A |
| 基金项目:国家自然科学基金(52162050)；光电技术与智能控制教育部重点实验室(兰州交通大学)开放课题(KFKT2020-11) |
|
| Fault diagnosis of small sample bearings based on fuzzy clustering and improved Densenet network |
|
WEI Wenjun1,2,ZHANG Xuanming1,YANG Liben1
|
|
(1.School of Automation and Electrical Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China; 2.Key Laboratory of Opto-Technology and Intelligent Control of Ministry of Education (Lanzhou Jiaotong University), Lanzhou 730070, China)
|
| Abstract: |
| In practice, the scarcity of failure data for bearings makes it challenging to meet the extensive training requirements of deep learning models. This paper leverages the fine-grained feature extraction capabilities of Convolutional Neural Networks and the classification ability of fuzzy clustering without the need for training, proposing a small-sample bearing fault diagnosis method based on fuzzy clustering and an improved Densenet network. Initially, the pretrained Densenet network is modified by removing the classification layer and retaining only the feature extraction layers, and designing an Adaptive Global Average Pooling (GAP) layer to replace the Fully Connected (FC) layer. Subsequently, fuzzy clustering is utilized instead of the Densenet networks softmax classification layer, eliminating the need for training to achieve classification. Experimental results demonstrate that by training the GAP layer parameters with small-sample data, the model significantly reduces the requirement for training samples. During diagnosis, bearing time-domain images are input into the network, outputting 1 920 feature data at the GAP layer. Feature vectors matrices are constructed from the feature data of different fault states. Fuzzy similarity matrices and fuzzy equivalence matrices are obtained using fuzzy clustering methods. As the confidence factor changes from high to low, dynamic clustering diagrams are derived from the corresponding Boolean matrices, thereby achieving bearing fault classification. |
| Key words: small sample global mean pooling layer transfer learning fuzzy clustering fault diagnosis |
