基于贝叶斯优化多尺度DenseNet的离心泵声信号故障诊断方法

打开文本图片集

中图分类号：TP183；TP206DOI：10.3969/j.issn.1004-132X.2025.09.015

Acoustic Signal Fault Diagnosis Method of Centrifugal Pumps Based on Bayesian Optimization MultiscaleDenseNet

CHEN Jian1，2 YAN Minghui1，² CHEN Pin1.2 1.Institute of Noise and Vibration Engineering，Hefei University of Technology，Hefei， 2.Automotive NVH Engineering &. Technology Research Center Anhui Province，Hefei，

Abstract：Since one-dimensional feature vectors might not retain temporal feature information，but neural networks had good effects on image recognition，an image data set constructed by fault sound signals of centrifugal pumps was used to conduct centrifugal pump fault diagnosis. A Bayesian optimized multiscale DenseNet fault diagnosis method was proposed for centrifugal pump sound signals. One-dimensional time series acoustic signals were transformed into two-dimensional image through Gram angle field，and the time information and fault characteristics were preserved. Then multiscale dense blocks were used to extract image features to enhance image feature reuse. The dropout layer and L2 regularization method were used to prevent overfiting，and Bayesian optimization algorithm was adopted to determine neural network hyperparameters. Finall，experimental verification was performed using centrifugal pump acoustic signals，and comparisons were made with other diagnostic methods. The results show that the Bayesian optimization multiscale DenseNet diagnosis model has a fault recognition rate of 99.5% for the test set.

Key words：centrifugal pump；fault diagnosis；Gram angle field；Bayesian optimization；multiscale DenseNet

0 引言

离心泵在石油化工、能源电力、矿山及国防等领域应用广泛。（剩余9440字）