基于DnCNN声音增强的高坝泄流微弱空化声音信号识别与提取

摘要
图/表
参考文献(26)
相关文章 (15)

全文: PDF (3156 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要空化空蚀是水工建筑物泄洪安全监测的重要内容，但是高坝泄流期间产生的强泄流噪声会大幅减弱空化空蚀音频监测方法的效果甚至致其失效。针对该问题提出了基于降噪卷积神经网络DnCNN（Denoising Convolutional Neural Network）声音增强的空化声信号增强方法，该方法依据语音增强思想，通过DnCNN实现带噪音频监测信号中空化声信号的增强。首先对该方法的实现原理和DnCNN网络结构进行了阐述，然后使用采集自空蚀和泄流实验的空化声信号和泄流噪声对该方法的效果进行验证，最后通过支持向量机信号多分类识别实验和单分类支持向量机空化声信号单分类识别实验对该方法的泛化性能和工程实用性进行评价。研究结果表明该方法能够有效提升带噪空化声信号的信噪比，极大地还原空化声信号的频谱结构特征，实现强泄流噪声中微弱空化声信号的识别与提取，同时该方法具有较强的泛化性能和较好的工程实用性。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	刘昉1
	王润喜1
	庞博慧2
	练继建1
	梁超1

关键词 ： DnCNN, 声音增强, 空化噪声, 支持向量机, 单分类支持向量机, 信号识别

Abstract：Cavitation and cavitation erosion is an important part of flood discharge safety monitoring of hydraulic structures, but the strong discharge noise generated during discharge of high dam will greatly reduce the effect of acoustic monitoring method of cavitation and cavitation erosion and even cause it to fail. To solve this problem, a method of cavitation acoustic signal enhancement based on Denoising Convolutional Neural Network (DnCNN) is proposed. This method enhances the cavitation acoustic signal in the monitoring signal with noise frequency through DnCNN according to the idea of speech enhancement. First, the principle of realization and DnCNN network structure of this method are elaborated, then the effect of this method is verified by cavitation noise and discharge noise collected from cavitation and discharge experiments. Finally, the generalization performance and engineering practicability of this method are evaluated by support vector machine signal multi-classification experiment and one-class support vector machine cavitation acoustic signal one-class classification experiment. The research results show that the method can effectively improve the signal-to-noise ratio of the noisy cavitation signal, greatly restore the spectrum structure characteristics of the cavitation signal, and realize the recognition and extraction of weak cavitation acoustic signal in strong discharge noise. At the same time, this method has strong generalization performance and better engineering practicability.

Key words： DnCNN acoustic enhancement cavitation noise support vector machine one-class support vector machine signal recognition

收稿日期: 2022-08-18 出版日期: 2023-11-15

引用本文:

刘昉1，王润喜1，庞博慧2，练继建1，梁超1. 基于DnCNN声音增强的高坝泄流微弱空化声音信号识别与提取[J]. 振动与冲击, 2023, 42(21): 52-62.
LIU Fang1, WANG Runxi1, PANG Bohui2, LIAN Jijian1, LIANG Chao1. Recognition and extraction of weak cavitation sound signals from high dam discharge based on DnCNN sound enhancement. JOURNAL OF VIBRATION AND SHOCK, 2023, 42(21): 52-62.

链接本文:

http://jvs.sjtu.edu.cn/CN/ 或 http://jvs.sjtu.edu.cn/CN/Y2023/V42/I21/52

[1] 龚波,袁寿其,骆寅,等. 叶轮空蚀状态下离心泵振动特性分析[J]. 振动与冲击, 2020, 39(2): 92-99.
GONG Bo, YUAN Shouqi, LUO Yin, et al. Vibration signal characteristics of centrifugal pumps with cavitation erosion impellers[J]. Journal of Vibration and Shock, 2020, 39(2): 92-99.
[2] 于安,王逸夫,雷婷婷. 绕振动水翼空化发展及水动力学特性研究[J]. 振动与冲击, 2022, 41(13): 265-274.
YU An, WANG Yifu, LEI Tingting. Cavitation development and hydrodynamic characteristics around oscillating hydrofoil[J]. Journal of Vibration and Shock, 2022, 41(13): 265-274.
[3] 冯永祥. 二滩水电站泄洪洞侧墙掺气减蚀研究[D]. 天津大学, 2008.
FENG Yongxiang. Research on aeration in cavitation-protection for the sidewalls of spillway tunnel in Ertan Hydropower Station[D]. Tianjin University, 2008.
[4] 戚定满,鲁传敬,何友声. 空化噪声谱特性研究[J]. 振动与冲击, 1999, (3): 34-38, 98.
QI Dingman, LU Chuanjing, HE Yousheng. An investigation on the spectra of bubble noise[J]. Journal of Vibration and Shock, 1999, (3): 34-38, 98.
[5] 阎兆立,刘进,程晓斌. 空化噪声分析及其在空化检测中的应用[C]//第十四届船舶水下噪声学术讨论会论文集, 2013: 584-588.
YAN Zhaoli, LIU Jin, CHENG Xiaobin. Cavitation noise analysis and its application in cavitation detection[C]// Paper Collection of the 14th Symposium on Underwater Noise of Ships, 2013: 584-588.
[6] 黄继汤. 空化与空蚀的原理及应用[M]. 北京: 清华大学出版社, 1991: 77-78.
HUANG Jishang. Principle and application of cavitation and cavitation erosion[M]. Beijing: Tsinghua University Press, 1991: 77-78.
[7] 商三英. 三门峡大坝双层泄水孔原型噪声观测与底孔改建设计[J]. 水利水电技术, 1995, (3): 5-12.
SHANG Sanying. Prototype observation of noise level in the two-layer discharge outlet and design on reconstruction of bottom outlet for Sanmenxia Dam[J]. Water Resources and Hydropower Engineering, 1995, (3): 5-12.
[8] 段文刚,侯冬梅,王才欢,等. 三峡大坝泄水建筑物水力学原型观测与分析[J]. 水利学报, 2019, 50(11): 1339-1349.
DUAN Wengang, HOU Dongmei, WANG Caihuan, et al. Hydraulic prototype observation and analysis of Three Gorges Dam discharge structures[J]. Journal of Hydraulic Engineering, 2019, 50(11): 1339-1349.
[9] 徐勇. 基于深层神经网络的语音增强方法研究[D]. 中国科学技术大学, 2015.
XU Yong. Research on deep neural network based speech enhancement[D]. University of Science and Technology of China, 2015.
[10] Boll S.. Suppression of acoustic noise in speech using spectral subtraction[J]. Ieee Transactions on Acoustics, Speech, and Signal Processing, 1979, 27(2).
[11] Xu Y, Du J, Dai L R, et al. An experimental study on speech enhancement based on deep neural networks[J]. Ieee Signal Process. Lett., 2014, 21(1).
[12] Xu Y, Du J, Dai L R, et al. A regression approach to speech enhancement based on deep neural networks[J]. Ieee/acm Transactions on Audio, Speech and Language Processing (taslp), 2015, 23(1).
[13] Du J, Huo Q. A speech enhancement approach using piecewise linear approximation of an explicit model of environmental distortions[C]//Interspeech, Conference of the International Speech Communication Association, Brisbane, Australia, September. DBLP, 2008.
[14] Zhang K, Zuo W M, Chen Y J, et al. Beyond a gaussian denoiser: Residual learning of deep CNN for image denoising[J]. Ieee Transactions on Image Processing : a Publication of the Ieee Signal Processing Society, 2017, 26(7).
[15] Zhao Y X, Li Y, Dong X T, et al. Low-frequency noise suppression method based on improved DnCNN in desert seismic data[J]. Ieee Geoscience and Remote Sensing Letters, 2019, 16(5).
[16] 王哲昊,简涛,王海鹏,等. 基于DnCNN的海面目标一维距离像识别方法[J]. 信号处理, 2021, 37(6): 932-940.
WANG Zhehao, JIAN Tao, WANG Haipeng, et al. One-dimensional range profile recognition method of sea-surface targets based on DnCNN[J]. Journal of Signal Processing, 2021, 37(6): 932-940.
[17] 钟铁,陈云,董新桐,等. 基于DBBCNN的沙漠区地震资料随机噪声衰减方法[J]. 石油地球物理勘探, 2022, 57(2): 268-278, 241-242.
ZHONG Tie, CHEN Yun, DONG Xintong, et al. Research on random noise attenuation method for seismic data from deserts based on DBBCNN[J]. Oil Geophysical Prospecting, 2022, 57(2): 268-278, 241-242.
[18] 张超铭,文晓涛,张晓琦,等. 基于DnCNN与约束卷积的地震数据去噪方法[EB/OL]. (2022-03-18)[2022-08-18]. https://kns.cnki.net/kcms/detail/11.2982.P.20220316.1806.032.html.
[19] Berouti M, Schwartz R, Makhoul J. Enhancement of speech corrupted by acoustic noise[C]// IEEE International Conference on Acoustics，Speech，and Signal Processing.Washington：IEEE，1979.
[20] Ioffe S, Szegedy C. Batch normalization: Accelerating deep network training by reducing internal covariate shift[C]// International Conference on International Conference on Machine Learning.PMLR, 2015: 448-456.
[21] He K M, Zhang X Y, Ren S P, et al. Deep residual learning for image recognition[C]//Proceedings of the Ieee Conference on Computer Vision and Pattern Recognition. 2016: 770-778.
[22] Vickers W, Milner B, Risch D, et al. Robust North Atlantic right whale detection using deep learning models for denoising[J]. The Journal of the Acoustical Society of America, 2021, 149(6).
[23] 罗先武,季斌,彭晓星,等. 空化基础理论及应用[M]. 北京: 清华大学出版社, 2020: 146.
LUO Xianwu, JI Bin, PENG Xiaoxing, et al. Basics of cavitation and its application[M]. Beijing: Tsinghua University Press, 2020: 146.
[24] 戚定满,沈焕庭. 小波在瞬态空化噪声分析中的应用[J]. 振动与冲击, 2001, (1): 84-86, 32, 100.
QI Dingman, SHEN Huanting. Wavelet analysis of cavitation noise[J]. Journal of Vibration and Shock, 2001, (1): 84-86, 32, 100.
[25] 蒲中奇,张伟,施克仁,等. 基于小波奇异性理论的水轮机空化检测[J]. 振动与冲击, 2005, (5): 77-79, 137.
PU Zhongqi, ZHANG Wei, SHI Keren, et al. Turbine cavitation testing based on wavelet singularity detection[J]. Journal of Vibration and Shock, 2005, (5): 77-79, 137.
[26] Schölkopf B, Williamson R C, Smola A, et al. Support vector method for novelty detection[J]. Advances in neural information processing systems, 1999, 12.