Voiceprint recognition model of hydropower unit rub-impact faults based on integrated EEMD-CNN

肖博屹; 曾云; 刀方; 邹屹东; 李想; 拜树芳

doi:10.11660/slfdxb.20240106

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

Voiceprint recognition model of hydropower unit rub-impact faults based on integrated EEMD-CNN

更新时间：2026-07-22

- Voiceprint recognition model of hydropower unit rub-impact faults based on integrated EEMD-CNN
- Journal of Hydroelectric Engineering Vol. 43, Issue 1, (2024)
- 作者机构：
- 作者简介：
- 基金信息：
- DOI：10.11660/slfdxb.20240106
  CLC：
- Online First：25 January 2024，
  
  Published：2024
- 稿件说明：
移动端阅览
肖博屹, 曾云, 刀方, et al. Voiceprint recognition model of hydropower unit rub-impact faults based on integrated EEMD-CNN[J]. Journal of Hydroelectric Engineering, 2024, 43(1).
DOI：

肖博屹, 曾云, 刀方, et al. Voiceprint recognition model of hydropower unit rub-impact faults based on integrated EEMD-CNN[J]. Journal of Hydroelectric Engineering, 2024, 43(1). DOI： 10.11660/slfdxb.20240106.

摘要

水电机组声纹信号包含大量反映内部机械状态的有效信息，为了准确提取水电机组磨碰故障声纹特征，提出一种基于聚合经验模态分解（EEMD）与卷积神经网络（CNN）相结合的水电机组磨碰声纹识别模型。首先将水电机组噪声信号进行EEMD分解，得到若干本征模态分量（IMF）和残余分量（Res），然后将得到的IMF和Res与原噪声信号构建融合特征向量；以融合特征向量为输入，碰磨故障输出，正常和碰磨故障试验数据为样本，训练CNN深度学习神经网络，得到水电机组磨碰故障识别器，识别水电机组磨碰故障。结合水机电耦合平台和实际机组试验磨碰数据，验证了所提方法对水电机组碰磨故障识别效果，平均准确率达到99.8%，且该方法识别效果显著优于其他几种识别模型。

Abstract

Hydroelectric unit voice signals contain a significant amount of valuable information reflecting their internal mechanical state. To accurately extract the voiceprint features of rubbing faults in hydroelectric units

this paper presents a hydroelectric unit rubbing fault voiceprint recognition model based on the fusion of Ensemble Empirical Mode Decomposition (EEMD) and Convolutional Neural Network (CNN). First

we use EEMD to decompose a noise signal from a hydroelectric unit into several Intrinsic Mode Functions (IMFs) and a residue component (Res); we use these IMFs and Res

along with the original signal

to construct a fusion feature vector. Then

the vector is used as an input to train a CNN deep learning neural network

with the normal and rubbing fault test data as samples

so as to obtain a rubbing fault recognizer for hydroelectric units. This new method is validated against the rubbing test data from both the hydro-mechanical coupling test stand and the in-situ experiment

with an average accuracy of 99.8%

demonstrating its performance superior to other recognition models for the rubbing faults of hydroelectric units.

关键词

Keywords

references

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Spectral subtraction combined with CNN for diagnosing cavitation faults in hydraulic turbines

Stacking-based deep ensemble model for concrete strength prediction considering aggregate gradation and derived features

Acoustic fault diagnosis of hydraulic turbines based on fusion of convolutional attention and WGAN-AE

Intelligent inspection technology based on 5G technology and its fault diagnosis application at hydropower stations

Fault diagnosis of hydroelectric sets based on SVD and DBN

Related Author

付必涛

何志宏

周鑫

李超顺

淮浩翔

肖龙

余红玲

吴斌平

Related Institution

School of Electrical Engineering, Xi’an University of Technology

State Grid Weinan Electric Power Company

State Grid Shaanxi Electric Power Research Institute

Institute of Water Resources & Hydropower Engineering, Xi’an University of Technology

Key Laboratory of Fluid Machinery and Power Engineering Equipment Technology in Hubei province,Wuhan University

AI问答

⁰