基于语义嵌入空间的离心泵未知复合故障识别

PDF(4824 KB)

振动与冲击 ›› 2024, Vol. 43 ›› Issue (13) : 61-69.

论文

南玲博1，陈帝伊1，张润强1，王田田1，黄卫宁2

作者信息 +

Unknown composite fault identification of centrifugal pump based on semantic embedding space

NAN Lingbo1, CHEN Diyi1, ZHANG Runqiang1, WANG Tiantian1, HUANG Weining2

Author information +

文章历史 +

摘要

针对离心泵各类型故障相互耦合难以诊断的问题，提出了一种语义嵌入空间的未知复合故障识别框架。首先采用深度学习与加权交叉熵损失建模来提取数据不平衡分布下的离心泵已知单一故障特征。接着采用零样本学习结合语义编码将单一故障属性构建成复合故障语义空间。最后通过全连接神经网络将样本特征映射到语义空间，并通过相似度测量来识别未知复合故障。在自吸和单级单吸离心泵两个数据集上进行验证，结果表明：在复合样本缺失的前提下，该方法对四组数据不平衡分布下的未知复合故障平均识别率为73.63%和82.25%，验证了所提方法的有效性和泛化性。

Abstract

An unknown compound fault identification framework based on semantic embedding space is proposed to solve the problem of coupling of various types of centrifugal pump faults. Firstly, deep learning and weighted cross entropy loss modeling are used to extract the known single fault features of centrifugal pump under imbalanced data distribution. Then zero shot learning combined with semantic encoding is used to construct a compound fault semantic space for a single fault attribute. Finally, a fully connected neural network is used to map the sample features to the semantic space, and the unknown compound faults are identified by similarity measurement. Two datasets of self-priming and single-stage single-suction centrifugal pump are validated. The results show that under the premise of missing compound samples, the average accuracy of unknown compound faults under the unbalanced distribution of four groups of data is 73.63% and 82.25%, which validates the effectiveness and generalization of the proposed method.

导出引用

南玲博1，陈帝伊1，张润强1，王田田1，黄卫宁2. 基于语义嵌入空间的离心泵未知复合故障识别[J]. 振动与冲击, 2024, 43(13): 61-69

NAN Lingbo1, CHEN Diyi1, ZHANG Runqiang1, WANG Tiantian1, HUANG Weining2. Unknown composite fault identification of centrifugal pump based on semantic embedding space[J]. Journal of Vibration and Shock, 2024, 43(13): 61-69

参考文献

[1] P M, Frank. Analytical and qualitative model-based fault diagnosis-a survey and some new results[J]. European Journal of Control, 1996, 2(1):6-28. [2] Rao A R, Dutta B K. Vibration analysis for detecting failure of compressor blade[J]. Engineering Failure Analysis, 2012, 25(10):211-218. [3] Li K, Zhang R, Li F C, et al. A new rotation machinery fault diagnosis method based on deep structure and sparse Least Squares Support Vector Machine[J]. IEEE Access, 2019, 7(99):26571-26580. [4] Sakthivel N R, Sugumaran V, Nair B B. Comparison of decision tree-fuzzy and rough set-fuzzy methods for fault categorization of mono-block centrifugal pump[J]. Mechanical Systems & Signal Processing, 2010, 24(6):1887-1906. [5] Bordoloi D J, Tiwari R. Identification of suction flow blockages and casing cavitations in centrifugal pumps by optimal support vector machine techniques[J]. Journal of the Brazilian Society of Mechanical Sciences and Engineering, 2017, 39(8): 2957-2968. [6] Farokhzad S, Ahmadi H, Jaefari A, et al. Artificial neural network based classification of faults in centrifugal water pump[J]. Journal of Vibroengineering, 2012, 14(4): 1734-1744. [7] Ranawat N S, Kankar P K, Miglani A. Fault diagnosis in centrifugal pump using support vector machine and artificial neural network[J]. Journal of Engineering Research, 2021, 9(8): 99-111. [8] Haggag S, Adly A R, Abdelaal MMZ. Artificial neural network model for fault diagnosis of rotating machine in ETRR-2 research reactor[J]. Arab Journal of Nuclear Sciences and Applications, 2022, 55(3): 55-61. [9] Wen L, Li X Y, Gao L, et al. A new convolutional neural network-based data-driven fault diagnosis method[J]. IEEE Transactions on Industrial Electronics, 2017, 65(7):5990-5998. [10] Manikandan S, Duraivelu K. Vibration-based fault diagnosis of broken impeller and mechanical seal failure in industrial mono-block centrifugal pumps using deep convolutional neural network[J]. Journal of Vibration Engineering & Technologies, 2022, 11(1):141-152. [11] Kumar D, Dewangan A, Tiwari R, et al. Identification of inlet pipe blockage level in centrifugal pump over a range of speeds by deep learning algorithm using multi-source data[J]. Measurement, 2021, 186(12):110146. [12] Hasan M J, Rai A, Ahmad Z, et al. A fault diagnosis framework for centrifugal pumps by scalogram based imaging and deep learning[J]. IEEE Access, 2021, 9(4): 58052-58066. [13] 刘飞,陈仁文,邢凯玲.基于迁移学习与深度残差网络的滚动轴承快速故障诊断算法[J].振动与冲击, 2022, 41(3):154-164. LIU Fei, CHEN Renwen, XING Kailing. Fast fault diagnosis algorithm for rolling bearing based on transfer learning and deep residual network[J]. Journal of Vibration and Shock, 2022, 41(3): 154-164. [14] Larochelle H, Erhan D, Bengio Y. Zero-data learning of new tasks [C]//Proceedings of AAAI Conference on Artificial Intelligence, Chicago, 2008. [15] Bishop, Christopher M. Pattern recognition and machine learning [M]. New York: Springer, 2006. [16] 郑非凡.基于ResNet深度神经网络的异常检测模型[J].网络新媒体技术, 2020, 9(2):16-22. ZHENG Feifan. Anomaly detection model based on ResNet deep neural network[J]. Journal of Network New Media, 2020, 9(2):16-22. [17] Lu C, Wang Y, Ragulskis M, et al. Fault diagnosis for rotating machinery: A method based on image processing[J]. Plos One, 2016, 11(10): 0164111. [18] 白欣田. 滚动轴承故障模拟实验系统设计及故障诊断方法研究[D].中国矿业大学，2021. BAI Xintian. Study on rolling bearing fault simulation experiment system design and diagnosis method[D]. China University of Mining and Technology, 2021. [19] 张景.典型故障状态下船用离心泵运行特性研究[D].江苏大学, 2017. ZHANG Jing. Research on performance of a marine centrifugal pump under typical malfunction situations[D]. Jiangsu University, 2017. [20] 聂建平. 基于支持向量机的离心泵故障诊断方法研究[D].哈尔滨工业大学, 2017. NIE Jianping. Research on fault diagnosis of centrifugal pumps based on support vector machine[D]. Harbin Institute of Technology, 2017. [21] 朱浩. 基于零样本学习的网络入侵检测研究与实现[D].上海师范大学，2021. ZHU Hao. The research and implementation of network intrusion detection based on zero-shot learning[D]. Shanghai Normal University, 2021. [22] 王望望,邓林峰,赵荣珍. 集成KPCA与t-SNE的滚动轴承故障特征提取方法[J].振动工程学报, 2021, 34(2): 421-430. WANG Wangwang, DENG Linfeng, ZHAO Rongzhen. Fault feature extraction of rolling bearing integrating KPCA and t-SNE[J]. Journal of Vibration Engineering, 2021, 34(2): 421-430.