Abstract：When using the one-dimensional vibration signal of a centrifugal pump to identify the cavitation state, there are problems such as inaccurate signal feature search and the signal being greatly affected by noise. A centrifugal pump cavitation fault diagnosis method based on image recognition is proposed. The savitzky-golay convolution smoothing algorithm is used to denoise the original vibration signal, and then the denoised signal is converted into a pseudo-RGB image. A convolution kernel is designed based on the characteristics displayed in the image, and the image is convolved to obtain the features picture. Single-channel transformation is performed on the feature picture to reduce data, and finally the single-channel image is used as the input of the LeNet-5 neural network for cavitation fault diagnosis. Experimental test results show that the method can accurately identify centrifugal pump cavitation faults while accelerating model training, and the accuracy can reach 1. The study provides a new method for rapid diagnosis of centrifugal pump cavitation.
WU D H, WU Z X, ZHOU P J, et al. Transient characteristics of cavitation in low specific speed centrifugal pumps［J］. Journal of Hydroelectric Engineering, 2018, 37(3): 96-105.
TIAN L Y, ZHANG Y Z. Fault diagnosis of submersible pump based on Local mean decomposition and fast independent component analysis［J］. Acta Metrologica Sinica, 2020, 41(5):585-591.
［5］
PANDA A K, RAPUR J S, TIWARI R. Prediction of flow blockages and impending cavitation in centrifugal pumps using support vector machine (SVM) algorithms based on vibration measurements［J］. Measurement, 2018, 130: 44-56.
CHEN J, HUANG K X, LV W Y, et al. Fault diagnosis method of variable working condition bearing based on VMD and convolutional neural network［J］. Acta Metrologica Sinica, 2021, 42(7): 892-897.
ZHENG H F, YU S S, WANG Y B, et al. Research on vibration tone system signal analysis method based on empirical mode decomposition and singular value Decomposition［J］. Acta Metrologica Sinica, 2016, 37(4):398-401.
［9］
SHERVANI-TABAR M T, ETTEFAGH M M, LOTFAN S, et al. Cavitation intensity monitoring in an axial flow pump based on vibration signals using multi-class support vector machine［J］. Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science, 2018, 232(17): 3013-3026.
MCKEE K K, FORBES G L, MAZHAR I, et al. A review of machinery diagnostics and prognostics implemented on a centrifugal pump［C］//Engineering Asset Management 2011: Proceedings of the Sixth World Congress on Engineering Asset Management. Cincinnati, USA, 2014: 593-614.
DAI T, ZHANG Y F, ZHANG K X, et al. The research progress of empirical mode decomposition and mode mixing elimination［J］. Application of Electronic Technique, 2019, 45(3): 7-12.
ZHANG X M, TANG J, HAN J. An EMD mode aliasing ekimination method based on SVD［J］. Noise and Vibration Control, 2016, 36(6): 142-147.
ALTOB M A S, BEVAN G, WALLACE P, et al. Fault diagnosis of a centrifugal pump using MLP-GABP and SVM with CWT［J］. Engineering Science and Technology, an International Journal, 2019, 22(3): 854-861.
［16］
LOTFAN S, SALEHPOUR N, ADIBAN H, et al. Bearing fault detection using fuzzy C-means and hybrid C-means-subtractive algorithms［C］//2015 IEEE International Conference on Fuzzy Systems . IEEE, Istanbul, Turkey,2015: 1-7.
［19］
WANG H, LI S, SONG L, et al. A novel convolutional neural network based fault recognition method via image fusion of multi-vibration-signals［J］. Computers in Industry, 2019, 105: 182-190.
FU X W, Gao X Q. Rolling bearing fault diagnosis method based on FDM and singular value difference spectrum［J］. Acta Metrologica Sinica, 2018, 39(5):688-692.
［17］
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: 58052-58066.
CHEN J, LIU Y Y, HUANG K X, et al. Rolling bearing fault diagnosis method based on singular value decomposition and independent component analysis［J］. Acta Metrologica Sinica,2022, 43(6): 777-785.
［23］
SAVITZKY A, GOLAY M J E. Smoothing and differentiation of data by simplified least squares procedures［J］. Analytical chemistry, 1964, 36(8): 1627-1639.
HUANG M H, LI M Y, DAI Y. Based on the Savitzky-Golay noise reduction algorithm of a deep foundation pit monitoring data［J］. Journal of Shantou University(Natural Science Edition), 2022, 37(2): 50-60.
JIANG Y H, JIAO W D, LI R Q, et al. Astudy on the method for eliminating mode mixing in B-spline empirical mode decomposition based on adaptive bandwidth constrained signal［J］. Journal of Vibration and Shock, 2018, 37(16): 83-90.
［18］
KUMAR A, GANDHI C P, ZHOU Y, et al. Improved deep convolution neural network (CNN) for the identification of defects in the centrifugal pump using acoustic images［J］. Applied Acoustics, 2020, 167: 107399.
［25］
OSHEA K, NASH R. An introduction to convolutional neural networks［EB/OL］. arXiv preprint arXiv: 1511, 08458, 2015.
WANG D, DING X J. Research on early mechanical weak fault diagnosis method based on envelope demodulated stochastic resonance and CEEMD.［J］. Acta Metrologica Sinica, 2016, 37(2):185-190.
HIDAYAT A Y, WIDODO A, HARYADI G D. Fault diagnostic system bearing centrifugal pump using K-Means method for thermography image and signal analysis vibrations［C］//MATEC Web of Conferences. EDP Sciences, Bali, Indonesia, 2018, 159: 02006.
CHEN J, SUN T H, HUANG K X, et al. Bearing fault diagnosis method based on histogram equalization and convolutional neural network［J］. Acta Metrologica Sinica, 2022, 43(7): 907-912.
YU Y W, ZHAO L L, ZHOU C C. Modified RMS calculation formulae for vehicle vibration responses and their engineering application［J］. Automotive Engineering, 2019, 41(9): 1088-1095.
LI H. Analysis of overfitting phenomenon based on deep learning［J］. China Science and Technology Information, 2020(14): 90-91.
［26］
LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition［J］. Proceedings of the IEEE, 1998, 86(11): 2278-2324.
［27］
LU Y, WANG Z, XIE R, et al. Bayesian optimized deep convolutional network for bearing diagnosis［J］. The International Journal of Advanced Manufacturing Technology, 2020, 108(1): 313-322.
2024, Vol. 45 
