统计对称噪声对叶片振声信号提取的探讨

针对统计对称噪声系统对称点噪声统计特性相对稳定的特点，提出了利用对称镜像点的信号降低噪声，克服不稳定因素影响的改进方法，数值仿真结果表明，该方法克服了非平衡变化的影响，提高了实时分析能力。     

噪声影响下的高准确度介损角测量方法研究

基于FFT的介损角检测方法易于硬件实现,但其准确度受频谱泄漏和噪声的影响较大.为此,提出了一种基于Black-man自卷积窗的高准确度介损角检测算法,建立了噪声影响下介损角检测结果的统计描述模型.首先采用Blackman自卷积窗对噪声影响下的畸变信号进行加权,其次利用离散频谱对称插值算法计算基波相位角,然后根据电压与电流基波相位差实现介损角检测,最后根据测量不确定度传播原理,推导了噪声对介损角检测准确度的统计模型,并通过仿真和实际检测2验证了本文算法和统计模型的正确性和可行性.     

小波神经网络在动态汽车衡称量数据处理中的应用

通过分析称重信号的数据波形图,发现数据当中掺杂着大量噪声信号,这在很大的程度上影响称重结果。通常使用的滤波方法能在一定范围内消除噪声信号,但是在路面不平、车辆振动的情况下,称量结果不稳定。针对这种情况提出了小波神经网络算法对称重数据信号进行去除噪声处理。实验仿真得出,利用小波神经网络算法对称重信号进行处理后,相对于传统的去噪滤波方法,能得到更理想的数据波形,使得称重结果与实际值的误差在±2%内。     

光电稳定跟踪装置中微机电陀螺应用研究

以微机电陀螺在高精度光电稳定跟踪装置中的应用为背景,研究了陀螺输出噪声对光电稳定跟踪平台精度的影响.结果表明,陀螺噪声会引起平台基准轴的抖动和缓慢漂移.根据微机电陀螺的实测数据,分析了其噪声特性.基于AR模型建立了微机电陀螺的噪声统计模型.研究了基于Kalman滤波的陀螺去噪算法,给出了去噪结果,分析了该算法不能够取得较好滤波效果的原因.针对Kalman滤波在微机电陀螺信号低频去噪方面的局限性,将基于阈值决策的小波去噪方法应用于微机电陀螺的信号处理中,给出了滤波结果.实测结果表明由于后者不依赖于噪声的精确模型,可根据噪声在不同频段的统计特性采用阈值决策滤波,具有更好的抑噪效果.最后给出了两种滤波算法的比较.     

车道线实时检测的形态学算法及其DSP实现

通过分析传统车道线检测方法易受强噪声影响的特点,提出了基于形态学的车道线实时检测算法,克服了传统算法缺点,算法简单、稳定、可靠;同时介绍了算法在DSP上的实现。经实际对分辨率为736×526的道路图像进行检测,该算法检测速度达到50Hz,完全满足实时性要求。     

基于鲁棒自适应SCKF的智能汽车目标状态跟踪研究

准确的自车和前车状态估计是智能汽车有效决策和控制的前提,而以往的研究通常不考虑噪声统计特性不确定的问题,导致某些情况下车辆状态估计的误差很大。为此,提出一种鲁棒自适应平方根容积卡尔曼滤波（Robust adaptive square-root cubature Kalman filter,RASCKF）算法,以降低噪声统计不确定性对估计精度的影响。首先,采用最大后验概率准则估计了过程噪声协方差和测量噪声协方差的统计值,以提高噪声稳定时状态估计的精确性。然后,基于标准化测量新息序列设计了故障检测规则,利用实时测量新息对噪声协方差进行校正处理,保证状态估计算法的鲁棒性。最后,在不同的噪声干扰工况下对RASCKF算法进行了仿真验证。结果表明,RASCKF算法在估计精度和稳定性上明显优于标准SCKF算法,有效地解决了智能汽车目标状态跟踪过程中噪声统计特性不确定的问题。     

基于改进单元损伤变量的结构损伤识别概率统计方法

针对测量噪声的存在严重影响了识别结果的稳定性和可靠性,选取改进后的单元损伤变量作为损伤识别指标与概率统计方法相结合,借助统计量和假设检验方法确定损伤判别临界值,得到检验的判错概率,消除测量噪声的不利影响,根据结构损伤前后参数统计值的变化,给出统计意义上的损伤识别结果。算例分析表明,基于改进单元损伤变量的结构损伤识别概率统计方法对噪声有很强的鲁棒性,能有效避免损伤误判的发生。     

DCT和相关分析在轴承故障诊断中的应用

针对滚动轴承故障诊断问题,提出一种在频域进行相关分析诊断故障的新方法。将振动信号进行离散余弦变换得到频域系数,对频域系数自动分段,对不同分段情况下的频域系数进行相关分析,比较不同分段情况下的相关系数,取其最大值,对应的频率即为故障频率。该方法克服了时域内相关分析受噪声影响导致诊断不准确的缺点,在信号含噪声情况下诊断结果稳定、准确。SKF 6205-2RS轴承诊断实例说明该方法的可用性。     

船用舵机两种液压驱动方式振动噪声对比试验

液压驱动是目前船用舵机的主要驱动方式,其振动噪声不仅会影响系统的平稳运行、人员的健康舒适,还会对设备和元件的寿命产生影响,因此,振动噪声水平是现代船舶装备设计的关键指标之一。针对双柱塞缸对称驱动及单柱塞缸非对称驱动典型船用舵机,通过容积控制和节流控制两种方式,进行了两种舵机系统振动噪声的对比试验,探讨了不同驱动方式、不同驱动速度等因素对船用舵机振动噪声的影响。结果表明:采用阀控双柱塞缸对称驱动可有效降低舵机系统的振动噪声。     

一种高信噪比的高阻抗测量处理电路

在高阻抗测量过程中,由于测量有效电流十分微弱,即使较小的外界噪声电流也极大影响测量信号的信噪比。在分析高阻抗测量中噪声电流的来源、噪声电流对信号的影响规律和噪声电流特点等的基础上,提出了一种简洁的消除噪声电流的方法,即高阻抗对称电阻电路的测量方法。这种方法主要运用差动处理的技术,消除由于外界电磁干扰引入到连接导线上的噪声电流。实验表明:这种处理电路的方法,不需要对测量电路采取过多的屏蔽处理措施,也适合高噪声空间电流干扰环境下高阻抗的测量。     

基于超声造影剂谐波图像的流场测量技术

本文研究了基于超声造影剂的血流运动场估计与显示技术,通过在血液中注入超声造影剂作为示踪粒子,对造影谐波图像进行时域相关处理,可得到成像部位的二维流场分布图。相对于常规Ｄｏｐｐｌｅｒ方法中用Ｄｏｐｐｌｅｒ回波信号的频偏计算流速值,该技术可直接从超声图像提取与夹角无关的流速矢量信息。本文通过流动模型验证该方法,浸入超声水槽中的乳胶管中流动着血液替代品,沿水流方向进行超声成像,对实验所得的造影后的Ｂ超图像以及谐波图像,用一种节省计算量的多尺度相关算法进行处理,并相互比较。结果显示,谐波图像相对Ｂ超基波图像具有更高的信杂比,从根本上解决了基波图像低信噪比对时域相关测量精度的限制,可以得到与夹角无关的二维血流场分布图,该方法是医学和有关工业领域中超声流场测量的一种有效的方法。     

A robust method for axis identification

The present paper proposes a new method for axis identification in discrete axially symmetrical geometric models. This method is based on-a-never-used-before property of the axially symmetrical surfaces for which the symmetry line of any section curve of the surface (or of a portion of it in the case of an incomplete axially symmetrical surface) always intersects the axis of symmetry of the surface. Thus the working principle of the method makes it very robust to local defectiveness, measurement noise and outliers.In order to compare it with the most cited methods presented in literature, several types of tests have been designed and performed. The robustness of those methods, on the one hand, has been evaluated by defining the Statistical Confidence Boundary at 1σ confidence level. The trueness of the method, on the other hand, has been evaluated on geometric models obtained by measuring real objects. The high robustness, which characterizes the proposed method, makes it particularly suitable for product geometric inspection where high accuracy is required.     

周期扫描磁信号特征对光磁复合方位 探测统计分布的影响

针对超近程来袭目标方位探测统计分布问题，研究了光磁复合方位测量方法周期扫描磁信号对探测精度的影响机理。建立了永磁体旋转扫描空间磁场数学模型，推导出周期扫描磁信号方程，结合磁信号特征和恒阈值时间测量方法，推导出周期扫描磁信号时间测量概率统计分布函数解析式。研究了扫描周期、磁场强度、阈值电压和噪声对时间测量概率统计分布的影响规律。结果表明，随着扫描周期和磁场强度的增加，概率分布函数对称性不受影响，概率分布半宽会随之减小，且分布峰值随之呈现0.75～1.88范围内的增加。随着阈值检测电压的提高，概率分布首先呈现半宽减小、峰值提升0.48态势，进而出现半宽增大、峰值降低0.54现象，且在峰值前后分布曲线的上升沿和下降沿出现不同走势。随着等效噪声电压的增加，概率分布函数对称性不受影响，但分布半宽会随之增大，且分布峰值随之减小0.4～0.48。     

基于形态奇异值分解和经验模态分解的滚动轴承故障特征提取方法

针对随机噪声和局部强干扰影响经验模态分解(Empirical mode decomposition,EMD)质量的问题,提出一种形态奇异值分解滤波消噪方法,并将其与EMD相结合形成一种新的故障特征提取方法。该方法首先对原始振动信号进行相空间重构和奇异值分解(Singular value decomposition,SVD),根据奇异值分布曲线确定降噪阶次进行SVD降噪,再形态滤波,最后把消噪后的信号进行EMD分解,利用本征模模态分量(Intrinsic mode function,IMF)提取故障特征信息。对仿真信号和实际轴承故障数据的应用分析表明,该方法能有效地提取轴承故障特征,诊断轴承故障,还可以减少EMD的分解层数和边界效应,提高EMD分解的时效性和精确度。     

改进的ESMD用于公共场所异常声音特征提取

由于公共场所异常声音的特殊性及背景噪声的复杂性,极点对称模态分解(ESMD)用于异常声音分解时,存在一些理论和技术上的缺陷。经分析认为公共场所异常声音为非线性、非平稳信号,背景噪声服从T分布。为此,提出改进的ESMD用于公共场所异常声音分解,得到有利于识别的特征。所提出方法的特点是将T分布噪声序列添加到具有背景噪声的异常声音信号中,以减小背景噪声对特征提取的影响;将模态分量的排列熵作为判定异常声音与背景噪声的准则,自适应筛选有效的模态分量;用对称中点插值法替代极值中点奇偶插值法,以缓解ESMD插值端点不明确带来的模态失真。在公共场所异常声音数据库上进行了相关实验。实验结果表明,所提出的方法与目前典型的时频信号处理方法相比,在提高公共场所异常声音分类识别率的同时,缩短异常声音的分解时间,是一种有效的公共场所异常声音特征提取方法。     

Enhanced wavelet-based methods for reducing complexity and calculation time in sonar measurements

We introduce two enhanced wavelet-based methods to improve accuracy and optimize calculation efficiency for sonar time delay estimation. Existing time delay estimations provide poor results because of many environmental effects such as noise, multipath, and crosstalk. While advanced digital signal processing (DSP) techniques have been proposed to overcome these problems, they entail increased complexity and calculation time. We use prediction in the methods we propose; the position where the reflected wave starts to occur is predicted by a recognition technique. The results are much better and the calculation time is shorter than other methods that use DSP techniques. In the first method, the optimization procedure is applied in the time domain, while in the second method, the optimization is calculated in the wavelet domain. Numerical comparisons and simulations using synthetic signals are provided to demonstrate the effectiveness of the proposed enhanced methods. We also demonstrate that our new algorithms are more stable than the existing ones and that the calculation time is reduced dramatically while maintaining increased accuracy, especially in a high-noise environment.     

无辨识预估PSD控制

针对原无辨识PSD的不足，本文提出了PSD参数的同步自适应方法，计算量小，且对于均值为零的测量噪声具有更强的抗噪能力。为了更好地应用于时滞过程，利用无时滞PSD系统渐近稳定的特点，提出了无预估模型的PSD自适应预估控制算法。仿真和实验验证了算法的有效性。     

Note: anti-strong-disturbance signal processing method of vortex flowmeter with two sensors

Some digital signal processing methods have been used to deal with the output signal of vortex flowmeter for extracting the flow rate frequency from the noisy output of vortex flow rate sensor and achieving the measurement of small flow rate. In applications, however, the power of noise is larger than that of flow rate sometimes. These strong disturbances are caused by pipe vibration mostly. Under this condition the previous digital signal processing methods will be unavailable. Therefore, an anti-strong-disturbance solution is studied for the vortex flowmeter with two sensors in this Note. In this solution, two piezoelectric sensors are installed in the vortex probe. One is called the flow rate sensor for measuring both the flow rate and vibration noise, and the other is called the vibration sensor for detecting the vibration noise and sensing the flow rate signal weakly at the same time. An anti-strong-disturbance signal processing method combining the frequency-domain substation algorithm with the frequency-variance calculation algorithm is proposed to identify the flow rate frequency. When the peak number of amplitude spectrum of the flow rate sensor is different from that of the vibration sensor, the frequency-domain subtraction algorithm will be adopted; when the peak number of amplitude spectrum of the flow rate sensor is the same as that of the vibration sensor, the frequency-variance calculation algorithm will be employed. The whole algorithm is implemented in real time by an ultralow power micro control unit (MCU) to meet requirements of process instrumentation. The experimental results show that this method can obtain the flow rate frequency correctly even if the power of the pipe vibration noise is larger than that of the vortex flow rate signal.     

Statistical modeling and signal reconstruction processing method of EMF for slurry flow measurement

According to the experimental data, an amplitude probability density function (PDF) model of the slurry flow signal is built up for electromagnetic flowmeter (EMF) by the method combining statistical analysis with numerical fitting, in order to reveal the effect of slurry noise on the flow signal and describe the features of slurry flow signal. Based on this model, a signal reconstruction processing algorithm is proposed to deal with the output signal of EMF sensor for realizing the slurry flow measurement. At the same time, the high-low voltage switching mode based square-wave excitation method is presented for EMF so as to reduce the slurry noise interferences. A slurry-type EMF transmitter is developed with a DSP chip – TMS320F28335, to implement the signal processing algorithm and control function. Finally water flow calibrations and slurry flow experiments are conducted to verify the reliability and stability of the method and system. Experimental results show that its measurement accuracy of water flow is better than 0.5%, and its steady-state volatility of paper slurry is less than 3%, and its dynamic response time is less than 4 s.     

基于支持矢量回归机的Hilbert-Huang变换端点效应问题的处理方法

Hilbert-Huang变换的端点效应表现在两个方面,对信号进行经验模态分解(Empirical mode decomposition, EMD)和对各个内禀模态函数(Intrinsic mode function,IMF)进行Hilbert变换时都会产生端点效应。为了克服 Hilbert-Huang变换中的端点效应,采用支持矢量回归机对信号延拓后再进行经验模态分解,该方法可以有效地克服EMD方法的端点效应问题,得到具有物理意义的内禀模态函数;然后再次采用支持矢量回归机对IMF分量进行延拓后进行Hilbert变换,可有效地抑制Hilbert变换中的端点效应,获得准确的瞬时频率和瞬时幅值,从而得到具有物理意义的Hilbert谱。对仿真和实际信号的分析结果表明,基于支持矢量回归机的数据序列延拓方法能有效地解决Hilbert-Huang变换中存在的端点效应问题,而且其效果优于基于神经网络的数据序列延拓方法。