基于Daubechies小波基的木材声学振动信号分析与评价

针对音板木材振动信号分析不深入的问题,本研究基于小波包分析方法对木材振动信号进行分析,建立一种木材声学振动评价的新方法。利用多通道FFT分析仪获取泡桐木材的振动信号,基于Daubechies小波基函数对振动信号进行3层小波包分解与重构,得到了振动信号的特征值,将小波包变换后木材振动信号的特征值与传统方法测定的声学振动性能参数进行比较分析。研究结果表明:木材的振动信号能量主要集中在0～1 000 Hz,在传播的过程中能量衰减迅速; 对振动信号进行时域分析时得出,木材的动态弹性模量随着峭度因子的增大而减小,随着峰值因子的增大而增大; 基于Daubechies小波基的分析方法得到的振动信号特征值与声学振动参数之间具有显著的相关性,其中木材信号的特征能量率与木材声学振动参数动态弹性模量、声辐射品质常数以及声传播速度之间存在显著的正相关关系,与声阻抗以及对数衰减率的相关性及变化趋势之间有显著的负相关关系。研究表明,可以应用Daubechies小波对木材振动信号进行分析与评价,此方法为木材声学振动性能的客观评价提供了新思路。

The previous studies on the analysis and processing of wood vibration signals were mostly limited to the fast Fourier transform, where advanced signal analysis methods were rarely used. Aiming at solving the problem that the soundboard wood vibration signal analysis is not in-depth and exploring the relationship between the characteristics of wood vibration signals and its acoustic vibration performance, a new method based on the wavelet packet analysis was established to analyze the wood vibration signals for wood acoustic vibration evaluation in this study. The vibration signal of Paulownia wood was obtained through the multi-channel fast Fourier transform analyzer. The signal was decomposed and reconstructed with a three-layer wavelet packet according to the Daubechies wavelet base, and the characteristic value of the vibration signal was attained. The characteristic value of the vibration signal after the wavelet packet transformation was compared and analyzed with the acoustic vibration performance parameters measured with traditional methods. The research results showed that the vibration signal energy of wood was mainly concentrated between 0-1 000 Hz, and the energy decayed rapidly during the propagation process. When analyzing the vibration signal in the time domain, it was found that the dynamic elastic modulus of wood decreased with the increase of the kurtosis factor, and increased with the increase of the crest factor. Moreover, it was revealed that there was a significant correlation between the eigenvalues of the vibration signal and the acoustic vibration parameters based on the Daubechies wavelet basis analysis method. There was a significant positive correlation among the characteristic energy rate of wood signals, the dynamic elastic modulus of wood acoustic vibration parameters, and the sound radiation quality constant. Among them, there was a significant positive correlation between the characteristic energy rate of the wood signal and the dynamic elastic modulus of the wood acoustic vibration parameters, the sound radiation quality constant, and the sound propagation velocity. There was a significant negative correlation between the correlation, and the change trend of acoustic impedance and logarithmic decay rate. The research showed that Daubechies wavelet could be used to process and analyze the wood vibration signals. It could extract the characteristic information of wood acoustic quality parameters based on the vibration signals. Furthermore, it could establish a connection with the existing acoustic parameter evaluation indicators to achieve the feasibility of obtaining wood acoustic vibration characteristics through the characteristic information of the signals. This method provides a new idea for the objective evaluation of wood acoustic vibration performance.

引言
1 材料与方法
2 结果与分析
3 结论

图1 木材共振频率测定试验原理<br/>Fig.1 Schematic diagram of the experimental principle of wood resonance frequency determination

图1 木材共振频率测定试验原理
Fig.1 Schematic diagram of the experimental principle of wood resonance frequency determination

图2 动态弹性模量与时域信号指标之间的关系<br/>Fig.2 The relationship between dynamic elastic modulus and time-domain signal indicators

图2 动态弹性模量与时域信号指标之间的关系
Fig.2 The relationship between dynamic elastic modulus and time-domain signal indicators

图3 泡桐木材振动信号的小波包重构系数图<br/>Fig.3 Wavelet packet reconstruction coefficient graph of Paulownia wood vibration signal

图3 泡桐木材振动信号的小波包重构系数图
Fig.3 Wavelet packet reconstruction coefficient graph of Paulownia wood vibration signal

图4 泡桐木材振动信号小波包分解后各层的能量率<br/>Fig.4 The energy rate of each layer of Paulownia wood vibration signal after wavelet packet decomposition

图4 泡桐木材振动信号小波包分解后各层的能量率
Fig.4 The energy rate of each layer of Paulownia wood vibration signal after wavelet packet decomposition

图5 小波包分解第1频带信号能量率与木材声学振动性能之间的关系<br/>Fig.5 The relationship between the energy rate of the first frequency band signal and the acoustic vibration performance of wood

图5 小波包分解第1频带信号能量率与木材声学振动性能之间的关系
Fig.5 The relationship between the energy rate of the first frequency band signal and the acoustic vibration performance of wood

图6 小波包分解第2频带信号能量率与木材声学振动性能之间的关系<br/>Fig.6 The relationship between the energy rate of the second frequency band signal and the acoustic vibration performance of wood

图6 小波包分解第2频带信号能量率与木材声学振动性能之间的关系
Fig.6 The relationship between the energy rate of the second frequency band signal and the acoustic vibration performance of wood

表1 小波包分解第一频带信号能量率与木材声学振动性能之间的回归方程及相关系数<br/>Table 1 The regression equations and correlation coefficients between the energy rate of the first frequency band signal and the acoustic vibration performance of wood

表1 小波包分解第一频带信号能量率与木材声学振动性能之间的回归方程及相关系数
Table 1 The regression equations and correlation coefficients between the energy rate of the first frequency band signal and the acoustic vibration performance of wood

表2 小波包分解第二频带信号能量率与木材声学振动性能之间的回归方程及相关系数<br/>Table 2 The regression equations and correlation coefficients between the energy rate of the second frequency band signal and the acoustic vibration performance of wood

表2 小波包分解第二频带信号能量率与木材声学振动性能之间的回归方程及相关系数
Table 2 The regression equations and correlation coefficients between the energy rate of the second frequency band signal and the acoustic vibration performance of wood

[1] 王成毓, 杨照林, 王鑫, 等. 木材功能化研究新进展[J]. 林业工程学报, 2019, 4(3): 10-18. DOI:10.13360/j.issn.2096-1359.2019.03.002.
[2] 李哲锋, 多化琼, 青龙. 电声乐器中木材声学振动性能对音响特性的影响[J]. 林业工程学报, 2018, 3(3): 18-23. DOI:10.13360/j.issn.2096-1359.2018.03.003.
[3] 吕少一, 刘美宏, 彭立民, 等. 木质材料在乐器与音箱领域的应用进展[J]. 木材工业, 2020, 34(1): 25-29. DOI:10.19455/j.mcgy.20200106.
[4] 刘嘉政, 王雪峰, 王甜. 基于深度学习的树种图像自动识别[J]. 南京林业大学学报(自然科学版), 2020, 44(1): 138-144. DOI:10.3969/j.issn.1000-2006.201809004.
[5] HASSAN K T S, TIPPNER J. Acoustic properties assessment of Neem(Azadirachta indica A. Juss.)wood from trees irrigated with secondarily treated wastewater[J]. BioResources, 2019, 14(2): 2919-2930. DOI:10.15376/biores.14.2.2919-2930.
[6] MIAO Y Y, ZHONG M L, LIU Z B, et al. Analysis of wood vibration energy attenuation based on FFT vibration signal[J]. BioResources, 2015, 10(1): 272-281. DOI:10.15376/biores.10.1.272-281.
[7] 岳孔, 程秀才, 贾翀, 等. 利用傅里叶变换红外光谱分析高温对木材-胶粘剂界面性能的影响[J]. 光谱学与光谱分析, 2019, 39(10): 3179-3183. DOI:10.3964/j.issn.1000-0593(2019)10-3179-05.
[8] 金潇男. 基于小波分析的不同类型振动特性研究[D]. 武汉: 长江科学院, 2015.
[9] 张曼, 王明泉, 杨顺民, 等. 基于小波分析的锂电池空耦超声信号降噪研究[J]. 测试技术学报, 2020, 34(4): 304-310. DOI:10.3969/j.issn.1671-7449.2020.04.005.
[10] 刘刚, 李祎宸, 王锴, 等. 基于小波包分析的含砂流中砂粒信号特征识别[J]. 石油机械, 2018, 46(5): 83-88. DOI:10.16082/j.cnki.issn.1001-4578.2018.05.014.
[11] 王二化, 颜鹏, 李昕, 等. 基于小波包能量分布特征的齿轮故障诊断方法研究[J]. 机床与液压, 2020, 48(1): 188-192. DOI:10.3969/j.issn.1001-3881.2020.01.39.
[12] 赵东, 朱红娟. 木材缺陷声发射信号的小波包分析处理[J]. 计算机工程与应用, 2010, 46(11): 220-222. DOI:10.3778/j. issn.1002-8331.2010.11.067.
[13] ZHU X D, LIU Y. Detection and location of defects in laminated veneer lumber by wavelet package analysis[J]. BioResources, 2014, 9(3): 4834-4843. DOI:10.15376/biores.9.3.4834-4843.
[14] 徐锋, 刘云飞. 胶合板加载损伤的声发射信号小波包特征提取[J]. 南京林业大学学报(自然科学版), 2011, 35(6): 101-106. DOI:10.3969/j.issn.1000-2006.2011.06.021.
[15] 董红平, 李明. 基于瞬时频率的木材声发射事件辨识与损伤监测[J]. 西北林学院学报, 2020, 35(2): 229-234. DOI:10.3969/j.issn.1001-7461.2020.02.35.
[16] 王鑫, 胡卫兵, 韩广森. 小波包能量谱在古木框架结构的损伤预警[J]. 工业建筑, 2013, 43(8): 24-28. DOI:10.7617/j.issn.1000-8993.2013.08.005.
[17] 杨扬. 基于泡桐木材振动特性的民族乐器声学品质预测模型研究[D]. 哈尔滨: 东北林业大学, 2017.
[18] 付宝刚, 李红义. 基于小波包能量分布和大小变化的结构损伤识别对比分析[J]. 中国建材科技, 2020, 29(4): 90-95.
[19] JIANG Y T, WU C X, DENG K F, et al. An audio fingerprinting extraction algorithm based on lifting wavelet packet and improved optimal-basis selection[J]. Multimedia Tools and Applications, 2019, 78(21): 30011-30025. DOI:10.1007/s11042-018-6802-y.
[20] STANKE L, KUBICEK J, VILIMEK D, et al. Towards to optimal wavelet denoising scheme-A novel spatial and volumetric mapping of wavelet-based biomedical data smoothing[J]. Sensors, 2020, 20(18): 5301. DOI:10.3390/s20185301.
[21] UMBRAJKAAR A M, KRISHNAMOORTHY A, DHUMALE R B. Vibration analysis of shaft misalignment using machine learning approach under variable load conditions[J]. Shock and Vibration, 2020, 2020(4): 1-12. DOI:10.1155/2020/1650270.
[22] ZHANG X, TANG H S, ZHOU D Y, et al. Experimental study on grout defects detection in precast shell wall based on wavelet packet analysis[J]. IOP Conference Series: Earth and Environmental Science, 2020, 510: 052090. DOI:10.1088/1755-1315/510/5/052090.
[23] TEIMOORI T, MAHMOUDI M. Damage detection in connections of steel moment resisting frames using proper orthogonal decomposition and wavelet transform[J]. Measurement, 2020, 166: 108188. DOI:10.1016/j.measurement.2020.108188.
[24] TUNG T H, WANG S H, HUANG C C, et al. Use of discrete wavelet transform to assess impedance fluctuations obtained from cellular micromotion[J]. Sensors, 2020, 20(11):3250. DOI:10.3390/s20113250.
[25] KUBOJIMA Y, OKANO T, MASAMITSU O. Effect of annual ring widths on structural and vibrational properties of wood[J]. Mokuzai Gakkaishi, 1997, 43(8): 634-641.
[26] CANIATO M, FAVRETTO S, BETTARELLO F, et al. Acousticcharacterization of resonance wood[J]. Acta Acustica United With Acustica, 2018, 104(6): 1030-1040. DOI:10.3813/aaa.919269.
[27] 陶鑫, 韩佳彤, 徐伟, 等. 云杉材纹理角度对钢琴音板声学振动性能的影响[J]. 木材工业, 2019, 33(4): 14-17. DOI:10.19455/j.mcgy.20190404.
[28] 秦丽丽, 苗媛媛, 刘镇波. 泡桐木材主要物理特征及化学组分对其声学振动性能的影响[J]. 森林工程, 2017, 33(4): 34-39. DOI:10.16270/j.cnki.slgc.2017.04.006.

备注

引言

1 材料与方法

2 结果与分析

3 结论

期刊信息

备注

引言

1 材料与方法

2 结果与分析

3 结 论

期刊信息

3 结论