基于循环频谱相干和DCNN的隔膜泵单向阀故障诊断方法研究

PDF(2831 KB)

振动与冲击 ›› 2021, Vol. 40 ›› Issue (14) : 237-244.

论文

冯泽仲1,2,3，熊新1,2,3，王晓东1,2,3

作者信息 +

Diaphragm pump check valve fault diagnosis method based on cyclic spectral coherence and DCNN

FENG Zezhong1,2,3,XIONG Xin1,2,3,WANG Xiaodong1,2,3

Author information +

文章历史 +

摘要

针对传统的机器学习方法过分依赖特征提取的质量，而深度学习在强干扰条件下其故障辨识率不佳的问题，提出了一种基于循环频谱相干(CSCoh)和深度卷积神经网络(DCNN)的故障诊断方法，并将其应用于实际工况环境下的隔膜泵单向阀故障诊断当中。对振动信号进行循环平稳特性分析，利用快速循环相关谱计算方法将原始振动信号生成二维CSCoh图；将生成的CSCoh图作为输入从而降低深度诊断模型中特征学习的难度，通过构建DCNN模型，并引入批量归一化和Dropout技术来提升模型的收敛速度和泛化能力；利用所提模型对故障进行分类识别，进而实现单向阀的故障诊断。结果表明，该方法可以准确地识别单向阀的故障类型，并具有较好的泛化性能。

Abstract

Aiming at the problem that traditional machine learning methods rely too much on the quality of feature extraction, and yet deep learning has poor fault recognition rate under strong interference conditions, a method based on cyclic spectral coherence (CSCoh) and deep convolutional neural network (DCNN) was proposed and applied to the fault diagnosis of a diaphragm pump check valve under the actual working conditions.Firstly, the cyclostationary characteristics of the vibration signal were analyzed, and the fast cycle correlation spectrum calculation method was used to generate a two-dimensional CSCoh map of the original vibration signal.Then, the generated CSCoh map was taken as an input to reduce the difficulty of feature learning in the deep diagnosis model.The DCNN model was constructed and batch normalization and Dropout technology were introduced to improve the convergence speed and generalization ability of the model.Finally, the proposed model was used to classify and identify the faults, and then realize the fault diagnosis of the check valve.The results show that the method can accurately identify the fault type of the one-way valve and has good generalization performance.

导出引用

冯泽仲，熊新，王晓东. 基于循环频谱相干和DCNN的隔膜泵单向阀故障诊断方法研究[J]. 振动与冲击, 2021, 40(14): 237-244

FENG Zezhong,XIONG Xin,WANG Xiaodong. Diaphragm pump check valve fault diagnosis method based on cyclic spectral coherence and DCNN[J]. Journal of Vibration and Shock, 2021, 40(14): 237-244

参考文献

［1］牟竹青, 黄国勇, 吴建德, 等.基于DEMD的高压隔膜泵单向阀早期故障诊断［J］.振动、测试与诊断, 2018, 38(4): 758-764.
MU Zhuqing, HUANG Guoyong, WU Jiande, et al.Early fault diagnosis of high pressure diaphragm pump check valve based on differential empirical mode decomposition ［J］.Journal of Vibration, Measurement & Diagnosis, 2018, 38(4): 758-764.

［2］MA J, WU J D, WANG X D.Fault diagnosis method of check valve based on multikernel cost-sensitive extreme learning machine ［J］.Complexity, 2017(6)： 1-19.

［3］YIN S, LI X W, GAO H J, et al.Data-based techniques focused on modern industry: an overview ［J］.IEEE Transactions on Industrial Electronics, 2014, 62(1): 657-667.

［4］岳应娟，王旭，蔡艳平.内燃机变分模态Rihaczek谱纹理特征识别诊断［J］.仪器仪表学报，2017，38(10):2437-2445.
YUE Yingjuan，WANG Xu，CAI Yanping.Internal combustion engine fault diagnosis based on identification of variational modal Rihaczek spectrum texture characterization ［J］.Chinese Journal of Scientific Instrument ，2017，38 (10): 2437-2445.

［5］LIU Y Y, YU Z W, ZENG M, et al.LLE for submersible plunger pump fault diagnosis via joint wavelet and SVD approach ［J］.Neurocomputing, 2016, 185:202-211.

［6］ZHAO X, JIA M.Fault diagnosis of rolling bearing based on feature reduction with global-local margin Fisher analysis ［J］.Neurocomputing, 2018, 315:447-464.

［7］LEI Y G, YANG B W, JIANG X W, et al.Applications of machine learning to machine fault diagnosis: a review and roadmap ［J］.Mechanical Systems and Signal Processing, 2020, 138:106587.

［8］文成林, 吕菲亚.基于深度学习的故障诊断方法综述［J］.电子与信息学报, 2020, 42(1): 234-248.
WEN Chenglin, L Feiya.Review on deep learning based fault diagnosis［J］.Journal of Electronics and Information Technology, 2020, 42(1): 234-248.

［9］JIAO J Y, ZHAO M, LIN J, et al.A comprehensive review on convolutional neural network in machine fault diagnosis ［J］.Neurocomputing, 2020, 417:36-63.

［10］高佳豪，郭瑜，伍星.基于SANC和一维卷积神经网络的齿轮箱轴承故障诊断［J］.振动与冲击, 2020, 39(19): 204-209.
GAO Jiahao, GUO Yu, WU Xing.Gearbox bearing fault diagnosis based on SANC and 1D CNN［J］.Journal of Vibration and Shock, 2020, 39(19): 204-209.

［11］李恒，张氢，秦仙蓉，等.基于短时傅里叶变换和卷积神经网络的轴承故障诊断方法［J］.振动与冲击, 2018, 37(19): 124-131.

LI Heng, ZHANG Qing, QIN Xianrong, et al.Fault diagnosis method for rolling bearings based on short-time Fourier transform and convolution neural network［J］.Journal of Vibration and Shock, 2018, 37(19): 124-131.

［12］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.

［13］CHEN Z Y, MAURICIO A, LI W H, et al.A deep learning method for bearing fault diagnosis based on cyclic spectral coherence and convolutional neural networks［J］.Mechanical Systems and Signal Processing, 2020, 140: 106683.

［14］ANTONI J. Cyclostationarity by examples ［J］.Mechanical Systems and Signal Processing, 2009, 23(4): 987-1036.

［15］DELVECCHIO S, D’ELIA G, DALPIAZ G.On the use of cyclostationary indicators in IC engine quality control by cold tests ［J］.Mechanical Systems and Signal Processing, 2015, 60:208-228.

［16］BORGHESANI P, SHAHRIAR M R.Cyclostationary analysis with logarithmic variance stabilisation ［J］.Mechanical Systems and Signal Processing, 2016, 70:51-72.

［17］ANTONI J, HANSON D.Detection of surface ships from interception of cyclostationary signature with the cyclic modulation coherence ［J］.IEEE Journal of Oceanic Engineering, 2012, 37(3): 478-493.

［18］ANTONI J. Cyclic spectral analysis in practice ［J］.Mechanical Systems and Signal Processing, 2007, 21(2): 597-630.

［19］ANTONI J, XIN G, HAMZAOUI N.Fast computation of the spectral correlation ［J］.Mechanical Systems and Signal Processing, 2017, 92:248-277.