PDF下载

[1]江新星,张雪琴,薛一冰,等.基于SOM-BP级联神经网络的电驱离心泵健康状态识别方法[J].油气储运,2025,44(03):1-15.
　JIANG Xinxing,ZHANG Xueqin,XUE Yibing,et al.Health status recognition method of electrically driven centrifugal pump based on cascade-structured SOM-BP neural networks[J].Oil & Gas Storage and Transportation,2025,44(03):1-15.

点击复制

基于SOM-BP级联神经网络的电驱离心泵健康状态识别方法

《油气储运》[ISSN:1000-8241/CN:13-1093/TE] 卷: 44 期数: 2025年03期页码: 1-15 栏目: 出版日期: 2025-03-25

Title:: Health status recognition method of electrically driven centrifugal pump based on cascade-structured SOM-BP neural networks

作者:: 江新星; 张雪琴; 薛一冰; 彭世亮

Author(s):: JIANG Xinxing; ZHANG Xueqin; XUE Yibing; PENG Shiliang

关键词:: 自组织映射; BP神经网络; 电驱离心泵; 健康状态识别

Keywords:: Self-Organizing Map (SOM); Back Propagation (BP) neural network; electrically driven centrifugal pump; health status recognition

分类号:: TE832

文献标志码:: A

摘要:: [目的]离心泵预测性维护是提升设备可靠性与运行效率的核心技术之一，在该过程中，对离心泵设备的健康状态识别是关键环节。然而，传统的健康状态识别方法多依赖于机器学习技术，高度依赖足量标记数据，难以直观清晰地表征监测数据与健康状态之间的对应关系，使其在实际复杂工况中的应用效果受限，亟需开发更精准、高效且适应性更强的健康状态识别方法。[方法]为此，提出了一种基于自组织映射（Self Organization Map, SOM）神经网络与BP（Back Propagation）神经网络级联的电驱离心泵健康状态识别方法。首先采用SOM神经网络方法对离心泵全生命周期振动数据进行预处理，提取时域、频域及时频域的多种统计特征与熵特征，从而全面表征设备的运行状态；其次，采用主成分分析法（Principal Component Analysis, PCA）对已提取的轴承振动信号特征进行降维与融合，有效减少冗余信息和计算复杂度，优化输入参数的模式，提升建模效率；最后，综合SOM神经网络与BP神经网络的优点，建立了基于SOM-BP级联神经网络的电驱离心泵健康状态识别模型。[结果]以某电驱离心泵的健康状态监测数据集为算例，对比了SOM-BP模型与常见的机器学习方法（随机森林、XG-boost）识别电驱离心泵健康状态的准确率，以R2、MSE、RMSE为模型评价指标，结果表明：基于SOM-BP级联神经网络模型的R2值、MSE值、RMSE值分别为0.901、0.8×10-6 m2/s4、9.12×10-4 m/s2，显著优于传统的机器学习方法，展现出良好的鲁棒性与适应性。[结论]所提方法不仅提升了离心泵健康状态识别的精度，还可为离心泵故障诊断与剩余寿命预测提供数据支撑，同时为其他旋转机械的健康状态管理与诊断提供了新思路。

Abstract:: [Objective] Predictive maintenance is recognized as one of the core approaches to improving the reliability and operational efficiency of centrifugal pumps. A crucial component of implementing this strategy lies in the health status recognition of these pumps. However, traditional methods for health status recognition primarily depend on machine learning techniques that require a substantial amount of labeled data. This dependency often leads to their limitations in effectively and clearly representing relationships between monitoring data and health status, thus constraining their application effect in real-world conditions, which are typically complex. Therefore, it is urgently needed to develop health status recognition methods that are more accurate, efficient, and adaptable. [Methods] This paper presents a health status recognition method for electrically driven centrifugal pumps, utilizing a cascade configuration of the Self-Organizing Map (SOM) neural network and the Back Propagation (BP) neural network. First, the SOM neural network was employed to preprocess the life-cycle vibration data of centrifugal pumps, focusing on extracting various statistical characteristics and entropy features in the time, frequency, and time-frequency domains, to fully characterize their operating states. Next, the extracted bearing vibration signal features were subjected to dimensionality reduction and integration using Principal Component Analysis (PCA). This facilitates the reduction of redundant information and computational complexity, the optimization of the pattern of input parameters, and the enhancement of modeling efficiency. Finally, a health status recognition model was established for electrically driven centrifugal pumps based on cascaded SOM-BP neural networks, leveraging the advantages of both neural networks. [Results] Based on the health status monitoring dataset of an electrically driven centrifugal pump, a comparison was conducted between the SOM-BP model and common machine learning techniques, such as random forest and XGBoost, to evaluate their accuracy in health status recognition of the pump, using R2, MSE, and RMSE as evaluation metrics. The results indicated that the R2, MSE, and RMSE values generated by the proposed model were 0.901, 0.8 × 10-6 m2/s4, and 9.12 × 10-4 m/s2, respectively. These results were significantly superior to those obtained from traditional machine learning methods, demonstrating enhanced robustness and adaptability. [Conclusion] The proposed method has demonstrated effectiveness in enhancing the accuracy of health status recognition for centrifugal pumps. Additionally, it provides valuable data support for fault diagnosis and residual life prediction of these pumps. The results of this study present a novel perspective on health status management and diagnosis that can be applied to other rotating machinery.

相似文献/References:

[1]赵会军,张青松,张国忠,等.基于BP神经网络顺序输送混油粘度预测[J].油气储运,2007,26(12):33.[doi:10.6047/j.issn.1000-8241.2007.12.009]
　ZHAO Huijun,ZHANG Qingsong.Study on Contamination Viscosity in Batch Transportation Based on the BP Neural Network[J].Oil & Gas Storage and Transportation,2007,26(03):33.[doi:10.6047/j.issn.1000-8241.2007.12.009]
[2]付道明,孙军,贺志刚,等.预测腐蚀管道剩余强度的新方法[J].油气储运,2004,23(4):12.[doi:10.6047/j.issn.1000-8241.2004.04.004]
　FU Daoming,SUN Jun.A New method to Predict Residual Strength of Corrosion Pipelines[J].Oil & Gas Storage and Transportation,2004,23(03):12.[doi:10.6047/j.issn.1000-8241.2004.04.004]
[3]喻西崇,赵金洲,纪录军,等.利用神经网络分析注水管道内腐蚀影响因素[J].油气储运,2003,22(2):27.[doi:10.6047/j.issn.1000-8241.2003.02.007]
　YU Xichong,ZHAO Jinzhou.Corrosion Influence Factors Analysis Using Neural Network for Injecting Pipeline[J].Oil & Gas Storage and Transportation,2003,22(03):27.[doi:10.6047/j.issn.1000-8241.2003.02.007]
[4]刘刚,周志龙,王正涛,等.改进BP神经网络在管道腐蚀速率预测中的应用[J].油气储运,2006,25(4):34.[doi:10.6047/j.issn.1000-8241.2006.04.009]
　LIU Gang,ZHOU Zhilong.The Application of the Improved BP Neural Network in Prediction of Pipeline Corrosion Rate[J].Oil & Gas Storage and Transportation,2006,25(03):34.[doi:10.6047/j.issn.1000-8241.2006.04.009]
[5]高山卜,钱成文,张沛,等.基于改进的BP神经网络管输能耗预测模型[J].油气储运,2014,33(8):869.[doi:10.6047/j.issn.1000-8241.2014.08.014]
　GAO Shanbu,QIAN Chengwen,ZHANG Pei,et al.Prediction model of pipeline energy consumption based on improved BP neural network[J].Oil & Gas Storage and Transportation,2014,33(03):869.[doi:10.6047/j.issn.1000-8241.2014.08.014]
[6]郑倩　张宏　刘啸奔.基于BP-MC方法的断层区管道的可靠性分析[J].油气储运,2019,38(12):1.
　ZHENG Qian,ZHANG Hong,LIU Xiaoben.Reliability analysis of pipeline in fault area based on BP-MC method[J].Oil & Gas Storage and Transportation,2019,38(03):1.
[7]徐长航,刘鹏谦,李振兴,等.基于声发射信号的气液两相流管道泄漏识别方法[J].油气储运,2021,40(10):1131.[doi:10.6047/j.issn.1000-8241.2021.10.007]
　XU Changhang,LIU Pengqian,LI Zhenxing,et al.Leakage identification method of gas-liquid two-phase flow pipeline based on acoustic emission signal[J].Oil & Gas Storage and Transportation,2021,40(03):1131.[doi:10.6047/j.issn.1000-8241.2021.10.007]
[8]肖荣鸽,刘博,王勤学,等.基于GRA-ABC-BPNN模型的城市燃气日负荷预测[J].油气储运,2022,41(08):987.[doi:10.6047/j.issn.1000-8241.2022.08.015]
　XIAO Rongge,LIU Bo,WANG Qinxue,et al.Daily load forecasting of urban gas based on GRA-ABC-BPNN model[J].Oil & Gas Storage and Transportation,2022,41(03):987.[doi:10.6047/j.issn.1000-8241.2022.08.015]
[9]孙天礼,韩雪,黄仕林,等.元坝高酸性气田地面管道内腐蚀预测[J].油气储运,2023,42(01):40.[doi:10.6047/j.issn.1000-8241.2023.01.006]
　SUN Tianli,HAN Xue,HUANG Shilin,et al.Internal corrosion prediction of ground pipeline of Yuanba high-sour gasfield[J].Oil & Gas Storage and Transportation,2023,42(03):40.[doi:10.6047/j.issn.1000-8241.2023.01.006]
[10]温泉,石小平,张利分,等.基于改进Tent映射的APSO-LSTM天然气需求量预测模型[J].油气储运,2023,42(06):702.[doi:10.6047/j.issn.1000-8241.2023.06.012]
　WEN Quan,SHI Xiaoping,ZHANG Lifen,et al.APSO-LSTM model based on improved Tent mapping for natural gas demand forecasting[J].Oil & Gas Storage and Transportation,2023,42(03):702.[doi:10.6047/j.issn.1000-8241.2023.06.012]

更新日期/Last Update: 2025-02-12