小波包和功率谱密度分析法在混凝土龄期强度监测中的应用

利用压电智能骨料激发的应力波,对混凝土28 d内目标龄期进行信号监测试验研究,并通过功率谱密度和小波包能量分析法对混凝土龄期内监测信号进行分析.试验结果表明:混凝土目标龄期内,随着时间的逐渐增加,混凝土试件中传感器监测信号的小波包能量和功率谱密度的幅值随时间的变化趋势呈现出非线性增长,6 d和12 d是增长变化的分界点,其幅值分别达到了2.345×104 V2、3.923×104 V2和61.54 pV2/Hz,98.86 pV2/Hz.其试验结果与混凝土试件在实际养护过程中试件强度的变化规律具有很好的一致性.     

基于压电陶瓷的不同服役年限木构件损伤监测

基于压电陶瓷的主动传感技术对不同服役年限的木构件进行损伤评估。取服役年限为0年、3年、40年的木构件制成三个尺寸相同的立方体试件,在每个试件的六个面上分别粘贴压电陶瓷片(PZT),用于信号的发射及接收。分别对不同服役年限试件传感器测量的信号进行了小波包和功率谱分析,发现应力波沿木材纵向传播时相较于服役年限短的试件,长年限试件的信号的幅值更大,沿木材径向及弦向传播时随木材服役年限的增加信号的幅值随之减小。结果表明,文中提出的方法能有效的识别不同服役年限的木构件,可为实际工程中不同服役年限木结构古建筑健康监测提供参考。     

基于能量特征的脑电信号特征提取与分类

为了快速、有效地提取脑电特征,提高分类正确率,采用带通滤波和小波包分析的方法提取Mu、Beta节律对应的脑电信号,在时域范围内,将信号幅度的平方作为能量特征值;在频域范围内,采用AR模型功率谱估计法所得的功率谱密度作为能量特征值.根据运动想象脑电信号特点,构造左右通道信号能量差值的符号特性作为分类判别依据,进行分类测试,方法简单.初步实验结果表明,所利用的两种方法的分类正确率达87.857%.     

信号检测系统中预处理滤波算法的研究

针对信号检测系统中原始信号采集后的预处理滤波问题,分别采用FIR滤波算法和LMS滤波算法对该问题加以分析和讨论。通过功率谱密度、小波能量系数分析评价体系,验证了预加重信号在能量谱提升方面起到的作用,同时得出LMS滤波算法在信号预处理方面优于FIR滤波算法的结论。最后将该预处理滤波算法信号检测系统应用于整体传感器信号的测试检测中,该实验取得了较好的效果。     

多周期信号的小波包方差分析方法

在多周期结构分析中,最大重叠离散小波变换得到的信号周期具有明显的局限性。在对比小波方差分析中,提出了用最大重叠离散小波包方差法分析不同尺度小波方差图、功率谱,从而得到信号周期估计的最大值。实验结果表明,对信号或时间序列周期结构的分析是一种有效的方法,该方法可以准确估计多周期信号的小波包方差。     

基于小波包能量谱的往复式隔膜泵故障诊断研究

针对往复式隔膜泵故障的多元性、不确定性和并发性的特点,提出了基于小波包能量谱的往复式隔膜泵故障诊断方法。小波包能将振动信号分解到不同子频带,通过各子频带信号的能量变化反映设备运行状况。通过采集往复式隔膜泵振动信号,进行小波包分解为多个子频带,求出各频带的能量和能量比例,然后对比故障振动信号和正常振动信号的频带能量谱比例图,找出发生故障的频带,进而找出往复式隔膜泵的故障特征频率,诊断出故障。实验表明:通过小波包能量谱对往复式隔膜泵进行故障诊断是有效可行的。     

小波包分析在刀具声发射信号特征提取中的应用

分析了刀具的切削状态,介绍了刀具的声发射信号检测系统和小波、小波包分析技术,以及小波包频带能量分解方法,提出了小波包分解功率监测特征量提取技术.通过在刀具声发射的一个实例信号中的应用,有效地区分了刀具的两种切削状态,验证了小波包分解功率监测特征量提取方法的可行性.     

基于小波包能量谱和M-ary SVM的功率变换器故障诊断

针对功率变换器的故障诊断问题,提出一种基于小波包能量谱和M-ary支持向量机的故障诊断方法。首先,通过小波包分解得到故障信号能量谱特征向量,并结合傅里叶变换分析故障信号主要频率特征点,实现故障特征向量的降维;然后,基于M-ary支持向量机的分类模型诊断出功率变换器多故障模式。实验结果表明,相比于传统的BP神经网络和一对一支持向量机故障诊断方法,本文方法诊断精度高,需要的子分类器数目少,诊断速度快,适用于在线故障诊断。      

木材缺陷声发射信号的小波包分析处理

在简要介绍小波包分析的分解和重构算法基础上,通过木材声发射实验采集声发射信号;利用小波包分析技术对三种不同缺陷类型的木材试件的原始数据进行消噪预处理,然后对信号进行分解,并对分解的信号进行小波包重构;运用“小波包-能量”法提取木材声发射信号特征值,实现了木材缺陷状态和声发射信号特征向量之间的映射关系。结果表明：用小波包分析确定木材缺陷程度是一种有效的方法。     

裂缝对埋入混凝土中压电陶瓷电-声特性的影响

利用压电陶瓷的电-声转换特性,将压电陶瓷圆片作为超声源埋入混凝土结构中,构成了压电埋入式混凝土机敏模块.该模块弥补了传统无损检测无法对混凝土结构进行长时间检测的不足.但由于混凝土内部会存在裂缝的影响,压电埋入式混凝土无损检测机敏模块的接收信号较弱.为了提高接收信号的强度,得到成分丰富的频谱,以分析裂缝对埋入混凝土中压电陶瓷电-声特性的影响,设计一种高压脉冲发生模块,该模块可产生幅度为400 V的窄脉冲激励压电陶瓷圆片,获得带宽为50兆赫兹的超声波频谱.实验表明,在高压脉冲激励下,接收信号的幅度值会随着裂缝与压电陶瓷片之间距离的增加而减小;接收信号与激励信号之间的时延会随着裂缝与压电陶瓷片之间距离的增加而增加.     

CFRM应变传感系统稳定性研究

本文利用碳纤维导电水泥基复合材料(CFRM)的压敏特性制成应变传感器,并研究了CFRM应变传感器在养护期间和养护完成后的电阻稳定性。试验结果表明,CFRM应变传感器的电阻在养护初期增长迅速,然后逐渐变慢,养护结束后很长一段时间内持续呈线性增长;此外,CFRM的电阻对温度很敏感,在降温0.105℃/h的过程中,电阻的变化率为2.027Ω/h;在100 h左右时间里,两个应变传感器的电阻差基本稳定在750Ω左右,拟合电阻变化率约为单个应变传感器电阻变化率的1/60,从而证明带补偿的CFRM应变传感系统可以补偿掉龄期和温度对CFRM应变传感器的影响,测量结果具有较好的稳定性。     

Structure damage diagnosis using neural network and feature fusion

A structure damage diagnosis method combining the wavelet packet decomposition, multi-sensor feature fusion theory and neural network pattern classification was presented. Firstly, vibration signals gathered from sensors were decomposed using orthogonal wavelet. Secondly, the relative energy of decomposed frequency band was calculated. Thirdly, the input feature vectors of neural network classifier were built by fusing wavelet packet relative energy distribution of these sensors. Finally, with the trained classifier, damage diagnosis and assessment was realized. The result indicates that, a much more precise and reliable diagnosis information is obtained and the diagnosis accuracy is improved as well.     

基于小波与GBDT的无人机传感器故障诊断

相对于有人飞行器,确保无人机传感器的正常工作更为重要.针对无人机传感器的故障诊断,提出了一种将小波特征提取与梯度提升决策树(GBDT)算法相结合的故障诊断方法.采用基于多层小波包分解的特征提取方法,将小波包分解系数与频带能量熵组合构成特征向量,相比单一的能量特征提取方法,有效提升了故障的可分性.采用梯度提升的策略对弱分类器进行迭代优化和线性组合,构成强分类器,使故障分类精度得到显著提高.仿真结果表明,该方法能有效进行特征提取和故障类型识别,且有较高的诊断精度和较强的泛化能力.     

分形噪声中BPSK信号检测的小波谱相关方法研究

小波变换作为一种新兴时频分析方法,其小波谱表征了信号的时频特性。对信号的小波谱以及与魏格纳-维尔分布之间的关系进行了较为详细地讨论。根据时频分析的理论,把小波谱扩展到了小波谱相关域,提出了小波谱相关分析方法,并应用到分形噪声和BPSK信号的分析上,探讨了它们的相应特性。根据探讨的结果,提出了分形噪声中BPSK信号检测的小波谱相关方法,并对讨论的结果进行了计算机仿真。     

Urban built-up land change detection with road density and spectral information from multi-temporal Landsat TM data

In this article, Landsat TM images acquired during the same season from both 1984 and 1997 were analysed for urban built-up land change detection in Beijing, China, where great changes have taken place during the recent decades. To reduce the spectral confusion between urban 'built-up' and rural 'non built-up' land cover categories, we propose a new structural method based on road density combined with spectral bands for change detection. The road density represents one type of structural information while the multiple Landsat TM bands represent spectral information. Road density maps for both dates were produced using a gradient direction profile analysis (GDPA) algorithm and then integrated with spectral bands. Results from the spectral-structural postclassification comparison (SSPCC) and spectral-structural image differencing (SSID) methods were evaluated and compared with spectral-only change detection methods. The proposed SSPCC method greatly reduced spectral confusion and increased the accuracy of land cover classification compared with spectral classification, which in turn improved the change detection results. This article also shows that the SSID change detection result complemented spectral band differencing by detecting areas with greater structural changes, some of which were missed, by spectral band differencing.     

Intelligent diagnosis of structural cracks with optimized sensing network

This paper considers intelligent diagnosis of structural cracks emanating from rows of rivet holes in thin metallic plates using active sensing network. Lamb waves are generated using actuators and propagate across the plates and received by sensors. We extract an effective feature called energy ratio change from time domain signals using wavelet transform. Then we develop neural networks using this feature to diagnose health condition. The sensing network is optimized by developing a mixed integer programming model. The results show that our method can effectively detect cracks and determine their locations, and the number of sensors of the sensing network can be significantly reduced while keeping high diagnostic accuracy. Important insights are also obtained such as in which area the sensing network has the weakest diagnostic capability.     

基于概率神经网络的钛钢爆炸复合棒材拉剪性能分类

研究基于超声信号的机器学习方法在钛钢爆炸复合棒材拉剪性能分类中应用。提出基于超声信号特征值的概率神经网络（Probabilistic neural network, PNN）评估分类方法,首先获取120组工件样本的水浸超声检测全序列A扫信号,对该信号进行时域分析和改进的协方差功率谱密度估计,得到钛钢上结合层深度、上复合层的反射频率、频谱能量、下复合层的反射频率、频谱能量以及下表面二次反射波衰减等6种特征值作为PNN输入;然后进行拉伸试验得到拉剪强度值作为PNN输出;最后以96个样本特征信号和拉剪强度值建立分类训练模型,其余24个样本超声特征信号作为测试集,对这些样本的拉剪强度值进行分类预测。实验结果表明,连续24次预测准确率为94.35%。本文研究为实现钛钢爆炸复合棒材拉剪性能快速、全覆盖检验找到新思路。     

Epileptic seizure endorsement technique using DWT power spectrum

EEG signals play significant role in the study of mental disorders. Epilepsy is one of the major mental disorders and need significant technological support in the treatment. A method proposed here is an endorsement technique for epileptic seizures using electroencephalogram (EEG) signals captured using non-invasive method. The method uses power spectrum density and discrete wavelet transformation (DWT). The impact of power spectral analysis along with the usage of EEG characteristics in endorsement of epilepsy is addressed here. A publicly available EEG epileptic dataset is processed using FIR filters along with DWT. The power spectrum density and its average were compared with specific spectrum to get the results and were compared against the standard EEG signal frequency range. It is found that the usage of DWT is more accurate and reliable to process and classify the EEG data for epilepsy endorsement.

     

Wavelet filter for noise reduction and signal compression in an artificial nose

This work presents results of the use of a wavelet filter for noise reduction and data compression of signals generated by artificial nose sensors. To verify the performance of the wavelet analysis in the treatment of odor patterns, we compare two widely used artificial nose classifiers, multi-layer perceptron neural network and time delay neural network in the analysis of signals generated by eight conducting polymer sensors exposed to gases derived from the petroliferous industry.