基于峰度的电力变压器铁芯松动故障在线监测方法

本文提出一种对电力变压器铁芯松动故障进行在线监测和估计的方法。通过改变变压器铁芯的压紧力得到模拟松动故障情况下的铁芯振动信号,使用谱峰度方法验证并分析了变压器铁芯振动信号中高频分量的冲击特性,在此基础上采用小波变换对铁芯振动信号进行分解,对分解得到高频分量计算其峰度值并用以监测铁芯压紧力的减小趋势。对实验变压器不同测点的多组数据分析表明,谱峰度能有效检测铁芯振动信号中的较高频率冲击成分及铁芯压紧力发生较大变化的情况,铁芯振动信号小波域中高频成分的峰度值能够很好的反映出铁芯压紧力变化的趋势。本方法准确的表述了铁芯振动信号中与松动故障相关的统计特性,为电力变压器铁芯的故障监测提供了一种可行的方法。

舰船辐射噪声的响度和音色特征模型

首先分析模拟舰船辐射噪声的归一化功率和Moore响度变化趋势,进一步用海试数据验证Moore响度的周期性变化特点,分析Moore响度周期性变化的原因,提取Moore响度均值作为一维特征.然后分析研究音色中有效的谱质心特征,根据谱质心的数学定义推导出谱质心带宽、谱质心斜度以及谱质心峰度的数学公式,举例分析其物理含义,并提出基于六分段频带的谱质心、谱质心带宽、谱质心斜度以及谱质心峰度共24维的音色特征模型.将Moore响度特征模型和音色特征模型相结合,从组合的25维特征的均值和方差分布可知特征具有一定的可区分性.最后设计3层BP神经网络分类器,海试数据表明基于Moore响度和音色的特征模型提取方法是有效的,且有利于提高分类识别率.

基于SVD-VMD和SVM滚动轴承故障诊断方法

针对故障滚动轴承振动信号中含有干扰信号,难以准确提取出故障信息,提出了一种基于奇异值分解(SVD)、变分模态分解(VMD)、和支持向量机(SVM)的滚动轴承故障诊断方法。首先利用奇异值分解对信号进行处理,根据奇异值峰度差分谱来确定分解后重构矩阵的有效阶数,然后根据该有效阶数重构信号,对重构后的信号进行VMD分解,根据上述有效阶数确定分解的本征模态函数(IMF)分量的个数,从分解后的IMF分量中提取故障特征参数,将其作为支持向量机的输入参数进行故障诊断。最后采用合肥工业大学轴承试验机进行验证,并与直接进VMD分解及基于带通滤波器信号去噪的故障诊断方法进行对比,结果表明该方法能有效识别滚动轴承的故障类型,可用于滚动轴承故障诊断。

一种小波调制信号的识别方法

小波调制是一种全新的多载波通信调制方式,该文主要研究加性高斯白噪声信道条件下小波调制信号的识别。根据小波调制信号的特点,以信号高阶累积量与功率谱峰度为特征参数,采用分步识别的方法对小波调制信号进行了识别。仿真结果表明,该方法非常有效而且能够很好的抑制高斯噪声的干扰。

Vertical-to-horizontal response spectral ratio for offshore ground motions:Analysis and simplified design equation

In order to study the differences in vertical component between onshore and offshore motions,the vertical-to-horizontal peak ground acceleration ratio(V/H PGA ratio) and vertical-to-horizontal response spectral ratio(V/H) were investigated using the ground motion recordings from the K-NET network and the seafloor earthquake measuring system(SEMS).The results indicate that the vertical component of offshore motions is lower than that of onshore motions.The V/H PGA ratio of acceleration time histories at offshore stations is about 50%of the ratio at onshore stations.The V/H for offshore ground motions is lower than that for onshore motions,especially for periods less than 0.8 s.Furthermore,based on the results in statistical analysis for offshore recordings in the K-NET,the simplified V/H design equations for offshore motions in minor and moderate earthquakes are proposed for seismic analysis of offshore structures.

Peak Profile Analysis in High Field Asymmetric Wave Ion Mobility Spectrometry

High field asymmetric wave ion mobility spectrometry （FAIMS） is a powerful tool to detect and characterize gas-phase ions, while the unsolvable partial differential equation of ions moving in ion drift tube poses a big challenge to FAIMS spectral peak analysis. In this work, a universal and effective model of FAIMS spectral peak profile has been proposed by introducing ion trajectory and loss height. With this model, the influence of the structure of ion drift tube, dispersion voltages, compensation voltages, and carrier gas flow rate on the FAIMS spectral peak characteristics like peak shape, full width at half maximum and peak height is analyzed and discussed. The results show that the influence of different factors on the FAIMS spectral peak profile can be qualitatively described by the model which agrees with the experimental data.

Application of wavelet transform in-ray spectra analysis

The frequency distribution of different ingredients in-ray spectra,e.g.,photo-peak,fluctuations of counts and Compton region,is separately analyzed.After wavelet transform of-ray spectra,the wavelet coefficients of a photo-peak increase with transforming scales and these coefficients show direct proportion with intensity of peak at determinate scale.A novel algorithm based on wavelet transform is proposed and studied.The results indicate that most of the photo-peaks in multi-spectra can be determined accurately,the-rays energy and intensity of the peak can also be determined.This method has the prospect of being applied in on-line multi-spectra analysis in such fields as radioprotection and nuclear safety monitoring.

Effects of drying process of Yuanzhi (Radix Palygalae) on its bioactive ingredients

OBJECTIVE： To study the effects of the drying processing in terms operational parameters on the bio- active constituents of six YuanzhJ （Radix Palygalae） samples across China. METHODS： Six Yuanzhi （Radix Palygalae） samples were investigated using thermogravimetry analysis. The heating courses were set in two ways： the temperature-programmed process from room temperature to 150℃ ,and the constant-temperature course at 50℃, 70℃ and 90℃. RESULTS： The peak temperature of six Yuanzhi （Radix Palygalae） samples ranged from 78℃ to 88℃. The mass loss rate of Yuanzhi （Radix Palygalae） alcohol-soluble extract was significantly increased when heated at 90℃. Four types of bioactive ingredients were detected in volatile oils of Yuanzhi （Radix Palygalae） sample from Shanxi province by Gas Chromatography-mass spectrometry analysis. Fourier Transform Infrared Spectroscopy results showed that the drying temperature exerted a great influence on types and amount of ingredients of Yuanzhi （Radix Palygalae）. The kinetic study showed that the constant-temperature drying process of Yuanzhi Radix Palygalae） samples could be well de- scribed by the Page Model, especially for the drying process at 50~C, in which R2 and SD values were more than 0.98 and less than 0.04, respectively.The drying constant k of three Yuanzhi （Radix Palygalae） samples from Shanxi, Gansu and Shaanxi provinces in China was corresponding to the Arrhenius equation, and their activation energies were 28.07, 2.5.38 and 21.48 kJ/mol, respectively. CONCLUSION： The drying process of Yuanzhi （Radix Palygalae） was very important for bioactive ingredients improvement in Yuanzhi （Radix Palygalae）. Temperature was a thermodynamic property significantly affecting the process.