共查询到19条相似文献,搜索用时 171 毫秒
1.
针对传统Teager能量解调算子方法对电梯运行系统中存在的强背景噪声较为敏感的不足,提出了一种改进的能量解调算子方法;采用了B样条技术与传统Teager能量算子方法进行结合,其中建立的B样条曲线对信号进行插值起到滤波作用;然后再利用Teager能量算子对滤波信号进行转换;最后利用傅里叶变换得到转换信号的频谱图从而揭示故障特征;所提出的基于B样条插值的能量解调方法不仅保留了传统能量解调算法的优点,如较高的解调精度和优秀的时间分辨率等,并且可以在强噪声背景下提取出微弱轴承故障特征;经实验验证实现了提高强背景噪声下的轴承故障检测的性能,能够在故障退化的早期检测故障,满足了实际工况下故障诊断上的应用。 相似文献
2.
针对轴承早期故障特征淹没在噪声信号中难以提取的问题,提出了基于自相关能量算子解调的故障诊断方法。首先,对实测信号进行自相关变换处理,抑制信号中的噪声成分;其次,用能量算子解调法对信号的自相关函数进行解调,提取故障特征。通过实测信号分析、与Hilbert解调法的相比,表明所提方法的有效性和优越性。 相似文献
3.
由于交互式分割中Live-wire算法运算速度较慢和操作复杂的问题,该文提出了改进的IS分割算法,在其代价函数中,使用Canny算子代替Laplace算子,并加入了像素边界函数、内边界函数和外边界函数等约束条件,较大的提高了算法运算速度的同时也提高了精度。 相似文献
4.
基于实数编码的多算子演化遗传算法 总被引:4,自引:1,他引:4
提出一种新的基于实数编码的多算子演化遗传算法(RMEGA)。实验结果表明该算法能够有效地改善标准遗传算法(SGA)的性能,克服其早熟收敛、运算速度低和精度较差的缺点。 相似文献
5.
《计算机测量与控制》2014,(4)
提出了一种利用经验模态分解(EMD)与Teager能量算子解调算法(EOSA)相结合的滚动轴承故障诊断方法;故障轴承振动信号往往是多分量AM-FM信号,首先对滚动轴承内圈故障调制信号进行EMD分解,得到若干个不同特征时间尺度的内禀模态信号(IMF),然后利用Teager能量算子解调法计算各个内禀模态信号的瞬时幅值,对主要内禀模态信号的瞬时幅值包络谱进行分析,可提取轴承故障信号特征信息;实验分析结果表明,利用经验模态分解与EOSA相结合的方法,可有效提取6205-2RS型深沟球滚动轴承的内圈特征频率fi=162.2Hz及其倍频。 相似文献
6.
以排除扰动因素对系统特征时间序列的冲击,还系统真实面貌,增强系统特征时间序列预测建模精度为目的,根据新信息优先利用原理,在对序列缓冲算子理论和已有序列弱化缓冲算子研究的基础上,构造了一类新的时间序列弱化缓冲算子,并与已有的系列弱化算子进行了建模精度比较。研究结果表明这类新的序列弱化缓冲算子能显著提高小样本时间序列的预测精度。 相似文献
7.
为了提高短法珀腔干涉式微压传感器峰值解调法的精度,本文研究了基于法珀干涉的石墨烯光纤微压传感器的解调方法.利用FDTD Solutions光学仿真软件仿真模拟传感器的反射光谱,搭建了传感器测试系统.采用Savitzky-Golay卷积平滑滤波和洛伦兹局部拟合相结合的峰值解调法,并利用多峰解调法解调出腔长,分析压力和腔长的关系.实验结果表明:采用Savitzky-Golay卷积平滑滤波明显滤除噪声信号,洛伦兹拟合寻找波谷的精度优于高斯拟合寻找波谷的精度,在0~0.1 MPa范围,传感器灵敏度为73.766 nm/kPa. 相似文献
8.
9.
10.
基于LoG算子的边缘零交叉二值化方法 总被引:6,自引:0,他引:6
针对全局阈值法和局部阈值法的不足,提出一种基于LoG算子的边缘零交叉二值化方法,并通过实验分析各算法的二值化效果及运算速度。 相似文献
11.
将希尔伯特振动分解(HVD)方法应用于非平稳电力谐波分析。在剖析非平稳多谐波信号的希尔伯特变换瞬时频率表达式基础上,通过低通滤波器直接估计出幅值最大谐波分量的瞬时频率,由同步检测获得相应的幅值和初相角,通过迭代运算自适应地检测出各次时变谐波分量参数。根据电网谐波的特点,设计一种满足性能指标的F IR低通滤波器实现HVD算法的关键环节,并采用波形匹配自适应边界延拓法避免截断处理的边界效应和累积误差。仿真表明,该方法能准确检测出频率波动和幅值突变等暂态谐波并适用于稳态谐波分析,与小波变换相比,无需选取基函数且精度与运算效率更高。 相似文献
12.
13.
14.
15.
A new algorithm for reconstructing a signal from its wavelet transform modulus maxima is presented based on an iterative method for solutions to monotone operator equations in Hilbert spaces. The algorithm's convergence is proved. Numerical simulations for different types of signals are given. The results indicate that compared with Mallat's alternate projection method, the proposed algorithm is simpler, faster and more effective. 相似文献
16.
提出一种新的技术,它自适应地选取正则化参数以取得较理想的恢复效果.利用小波变换,分析正则化算子和正则化参数对图象残差的各子频段能量的影响.在本文条件下,我们论证正则化算子取拉普拉斯算子比取恒等算子恢复性能好,并且预测噪声能量.实验结果表明本文提出的方法不需要知道噪声能量,也能够自适应地确定正则化参数并且恢复性能比传统的方法好,恢复效果非常接近最优恢复. 相似文献
17.
《国际计算机数学杂志》2012,89(3):545-554
The traditional way to define the instantaneous frequency of a real-valued signal is based on the analytic signal obtained by means of the Hilbert transform. Vakman showed the justification of this method by proving that under some fundamental conditions the imaginary part signal operator ? must be the Hilbert transform. This paper shows that his arguments are not valid in the finite-energy signal space L 2(?) – an important space of signals. In this paper, we propose substitutive conditions and establish the L 2(?) version of Vakman's analysis, which is a meaningful supplement to Vakman's theory. 相似文献
18.
Jay L. Adams Tom T. Hartley Robert J. Veillette 《Computers & Mathematics with Applications》2010,59(5):1773-1781
In this paper, the Rayleigh–Ritz method of estimating the eigenvalues of an operator on a Hilbert space is utilized to determine the magnitude of the largest eigenvalue for the Hankel operator of fractional-order systems, the Hankel norm. This provides a measure of the possible retrievable energy from the system in the future compared to the energy that was put into the system in the past. The application of the Rayleigh–Ritz method to obtaining underestimates of the Hankel norm of a fractional-order system is described. Several examples are given, demonstrating the method. 相似文献
19.
The analysis of the numerical aspects of Hilbert transform spectroscopy based on the a.c. Josephson effect is presented. The resolving power of Hilbert transform spectroscopy is determined by such factors as the linewidth of the Josephson oscillations (intrinsic or natural resolution limit) and the limitation of the measurement interval (extrinsic or technical resolution limit) like in any spectroscopic technique based on some integral transformations. The deconvolution problem in Hilbert transform spectroscopy is posed and its solution is considered using the approach of the 1st kind integral equation for the spectrum of the incident radiation constructed from the input data of the Hilbert transform spectroscopy—the ‘hilbertogram’. The program package RECOVERY based on the maximum likelihood method is used for this purpose. This method allows to attain the maximum possible resolution enhancement in output result for a given signal-to-noise ratio in the input experimental data. The samples of numerical simulations and the spectrum of frequency-modulated BWO radiation measured by means of the Josephson junction made from high-Tc superconductor are presented. It is shown also that the integral equation approach allows to recover the sought spectrum beyond the intrinsic resolution limit and to achieve the superresolution. 相似文献