基于SVD-CWT和CNN的水轮发电机转子故障识别

水轮发电机转子振动故障识别是水电站运维的重难点问题,为此提出一种基于转子振动信号的故障识别方法。首先针对发电机转子的非平稳和非线性振动信号,采用奇异值分解(SVD)并结合能量差分谱理论进行降噪预处理;对预处理数据使用连续小波变换(CWT)转换为时频图并形成图像数据集;然后将该图像数据集作为卷积神经网络(CNN)输入,通过CNN多层池化及卷积形成分布式故障特征表达,最终实现发电机转子故障模式识别和分类。经实验验证,该方法准确率达到99.5%以上,能有效识别出发电机转子的故障类型。

基于CWT和改进YOLOv5s的港口机械轴承故障诊断策略分析

轴承故障诊断常用的方法是振动分析,分析和预处理收集的振动信号。然而,轴承的振动信号具有不均匀的特性,这使得提取特性特别困难。在此基础上,本文将CWT和改进的YOLOv5S结合港口机械轴承故障进行研究,然后研究基于CWT和改进的YOLOv5S港口机械轴承故障诊断策略,旨在为有关人员提供参考帮助。

一种基于优化VMD-CWT-CNN的柱塞泵配流盘磨损状态识别方法

为解决一维振动信号难以充分挖掘表达状态特征信息以及柱塞泵配流盘磨损早期识别问题,基于卷积神经网络(Convolutional neural networks,CNN)优秀的图像处理能力,提出了一个优化VMD-CWT-CNN模型。首先,采用连续小波变换(Continuous wavelet transform,CWT)对信号进行预处理,得到信号的二维时频图,作为CNN模型的一路输入,将状态识别问题转化为CNN图像识别问题。其次,基于相关系数对变分模态分解(Variational mode decomposition,VMD)参数优化后,利用优化VMD对振动信号进行预处理,再以相关系数和峭度值最大为优选原则,甄选出三组蕴含故障特征的本征模态函数(Intrinsic mode function,IMF),将其重组为三通道一维信号,作为CNN模型的另一路输入。最后,在CNN模型中将两路信息汇聚并得到柱塞泵配流盘磨损状态识别分类结果。实验中,此方法分别采用优化VMD和CWT对振动信号预处理,再结合CNN对磨损状态进行分类。实验结果表明,该方法对于配流盘磨损的三种状态的识别效果显著优于单路输入的CNN模型以及典型的深度学习方法和机器学习分类器。因此,优化的VMD-CWT-CNN方法可以更准确地实现柱塞泵配流盘磨损状态识别。

基于DT-CWT和SVD的变电站直流系统接地故障检测技术研究

变电站直流系统的状态直接关系到变电站的正常运行,为了对变电站直流系统出现的接地故障快速、准确定位,提出了一种双树复小波变换(DT-CWT)和奇异值分解(SVD)相结合的变电站直流系统接地故障检测新方法。该方法首先利用DT-CWT对支路电流信号进行分解来构建Hankel矩阵;然后对Hankel矩阵进行SVD分解,得到一系列奇异特征值;再次,利用相邻奇异值差值构建奇异值差分谱,通过奇异值差分谱最大峰值来保留有效的奇异值个数;最后,利用保留的奇异值来重构低频信号。算例分析结果表明,该方法能够准确地从支路电流信号中提取出低频交流信号,可以对变电站直流系统接地故障进行准确定位,很大程度上减小对地电容对检测精度的影响。

一种高效率MCU芯片Multi-Sites测试技术

介绍了使用Multi-Sites工程测试技术提高MCU芯片测试效率的方案。针对MCU芯片Multi-Sites测试难点,阐述了在MCU芯片Multi-Sites测试中电性能测试、功能测试的影响因素和解决方案,并对MCU芯片Multi-Sites测试过程中经常遇到的干扰因素进行分析,保证MCU芯片Multi-Sites测试获得稳定可靠的性能参数,有效提高测试效率。

基于Multi-Site并行测试的效率分析与研究

在晶圆芯片测试时,提高单位时间内的效率能够实现低投入、高产出的效果。在采用Multi-site方式的并行测试中,需要先解决选择何种产品进行并行测试,如何确定并行site数等问题,然后再用最高效率的方法设计确定Multi-site并行测试方案。从软件和硬件方面,分析当前流行的Multi-site并行测试的效率,研究了影响Multi-site并行测试效率的各种因素,并对其影响深度和范围进行分析,给出相应的对策和提高效率的解决方法,同时还提出了溢出die计算方法,通过选择适当的site数,减少无用touchdown次数,提高测试效率。

基于VMD-CWT-CNN的滚动轴承故障诊断

针对传统故障诊断方法需人工提取特征且识别率低的问题,提出了一种基于变分模态分解(Variational Modal Decomposition,VMD)和连续小波变换(Continuous Wavelet Transform,CWT)并结合卷积神经网络(Convolutional Neural Network,CNN)的VMD-CWT-CNN模型用于滚动轴承故障诊断。首先,通过VMD将轴承振动信号分解为多个具有不同中心频率的模态分量;其次,采用CWT将模态分量进行计算并转为二维时频图;最后,将时频图输入结构裁剪后的EfficientNet卷积神经网络自动提取特征并完成滚动轴承的故障诊断。使用提出的方法对美国凯斯西储大学10类轴承故障数据进行多次实验,平均准确率达到了99.86%,能有效地完成滚动轴承信号的特征提取以及损伤程度的精确诊断。

Nuclephilic ring opening of epoxides promoted by multi-site phase-transfer catalyst:An efficient and eco-friendly route to synthesis of β-hydroxythiocyanate

A highly effective and mild protocol for ring opening of epoxides with NH4SCN in the presence of catalytic amount of a multi- site phase-transfer catalyst, α,α＇,α＂-N-hexakis（triethylammoniummethylene chloride）-melamine, is developed. A variety of ^-hydroxy thiocyanates as important intermediates in agricultural and pharmaceutical chemistry were obtained in high yields with excellent regioselectivity and in short reaction times. 2009 Ali Reza Kiasat. Published by Elsevier B.V. on behalf of Chinese Chemical Society. All rights reserved.

Accurate Multi-Site Daily-Ahead Multi-Step PM_(2.5)Concentrations Forecasting Using Space-Shared CNN-LSTM

Accurate multi-step PM_(2.5)(particulate matter with diameters≤2.5 um)concentration prediction is critical for humankinds’health and air populationmanagement because it could provide strong evidence for decisionmaking.However,it is very challenging due to its randomness and variability.This paper proposed a novel method based on convolutional neural network(CNN)and long-short-term memory(LSTM)with a space-shared mechanism,named space-shared CNN-LSTM(SCNN-LSTM)for multi-site dailyahead multi-step PM_(2.5)forecasting with self-historical series.The proposed SCNN-LSTM contains multi-channel inputs,each channel corresponding to one-site historical PM_(2.5)concentration series.In which,CNN and LSTM are used to extract each site’s rich hidden feature representations in a stack mode.Especially,CNN is to extract the hidden short-time gap PM_(2.5)concentration patterns;LSTM is to mine the hidden features with long-time dependency.Each channel extracted features aremerged as the comprehensive features for future multi-step PM_(2.5)concentration forecasting.Besides,the space-shared mechanism is implemented by multi-loss functions to achieve space information sharing.Therefore,the final features are the fusion of short-time gap,long-time dependency,and space information,which enables forecasting more accurately.To validate the proposed method’s effectiveness,the authors designed,trained,and compared it with various leading methods in terms of RMSE,MAE,MAPE,and R^(2)on four real-word PM_(2.5)data sets in Seoul,South Korea.The massive experiments proved that the proposed method could accurately forecast multi-site multi-step PM_(2.5)concentration only using self-historical PM_(2.5)concentration time series and running once.Specifically,the proposed method obtained averaged RMSE of 8.05,MAE of 5.04,MAPE of 23.96%,and R^(2)of 0.7 for four-site daily ahead 10-hourPM_(2.5)concentration forecasting.

Current climate overrides past climate change in explaining multi-site beta diversity of Lauraceae species in China

Background:We aimed to characterise the geographical distribution of Sørensen-based multi-site dissimilarity(β_(sor))and its underlying true turnover(β_(sim))and nestedness(βsne)components for Chinese Lauraceae and to analyse their relationships to current climate and past climate change.Methods:We used ensembles of small models(ESMs)to map the current distributions of 353 Lauraceae species in China and calculated β_(sor) and its β_(sim) and β_(sne) components.We tested the relationship between β_(sor),β_(sne) and β_(sim) with current climate and past climate change related predictors using a series of simultaneous autoregressive(SAR_(err))models.Results:Spatial distribution of β_(sor)of Lauraceae is positively correlated with latitude,showing an inverse rela-tionship to the latitudinalα-diversity(species richness)gradient.High β_(sor) occurs at the boundaries of the warm temperate and subtropical zones and at the Qinghai-Tibet Plateau due to high β_(sne).The optimized SAR_(err) model explainsβ_(sor) andβ_(sne) well,but notβ_(sim).Current mean annual temperature determinesβ_(sor) and β_(sne) of Lauraceae more than anomalies and velocities of temperature or precipitation since the Last Glacial Maximum.Conclusions:Current low temperatures and high climatic heterogeneity are the main factors explaining the high multi-siteβ-diversity of Lauraceae.In contrast to analyses of the β-diversity of entire species assemblages,studies of single plant families can provide complementary insights into the drivers of β-diversity of evolutionarily more narrowly defined entities.

Preparation of Cellulose Nanofibrils by Multi-Site Regioselective Oxidation

Cellulose nanofibrils(CNFs)are promising sustainable materials that can be applied to nanocomposites,as well as medical and life-sciences devices.However,methods for the preparation of these important materials are energy intensive because heating and mechanical disintegration are required to produce cellulose fibers below 100 nm in size.In this study,CNFs were prepared through the multi-site regioselective oxidation of cellulose with 2,2,6,6-tetramethylpiperidine-1-oxyl(TEMPO)and periodate at room temperature(20–25°C),without any mechanical-disintegration treatment.Transmission electron microscopy(TEM)revealed that the CNFs had the average widths of 14.1,55.4,and 81.9 nm for three different treatments.Fourier-transform infrared spectroscopy revealed that carboxyl groups were created on the surfaces of the microfibrils,while X-ray diffraction studies showed that the cellulose I structure was maintained after oxidation,and that the cellulose nanofibril crystallinity index exceeded 70%.These results demonstrate that CNFs can be prepared by multi-site regioselective oxidation at room temperature in the absence of mechanical disintegration.In addition,a model was developed to calculate the total content of carboxylate and aldehyde groups of CNFs prepared by the TEMPO mediate oxidation,the periodate oxidation,and the multi-site regioselective oxidation methods based on the particle width determined by TEM.The calculated values of the model were in good agreement with the total content(experimental value)of carboxylate and aldehyde groups of CNFs prepared by the TEMPO-mediated oxidation and the multi-site regioselective oxidation methods.However,the model was not valid for CNFs prepared by the periodate oxidation method.

Multi-site spinal stimulation strategies to enhance locomotion after paralysis

With the advent of spinal cord epidural stimulation techniques,i.e.,electrically enabled motor control（e Emc）in combination with activity dependent locomotor training,humans with traumatic complete sensori-motor paralysis are able to initiate voluntary leg movements and achieve gains in postural control,and bladder and sexual function（Angeli et al.,2014）.

基于ATE的电源芯片Multi-Site测试设计与实现

介绍了电源芯片的多Site测试设计与实现。基于CTA8280测试系统,通过对芯片CP(晶圆测试)要求进行分析,设计了8 Site测试电路外围,能够实现对晶圆进行8 Die并行测试。测试结果显示,该方案能够有效提升该电源芯片的测试效率,降低测试成本。

ON ANALYSIS METHOD FOR INFRARED MULTI-SITES SYSTEM PERFORMANCE

The calculation method about infrared multi-sites passive system location is introduced based on the principle of the weighted least square method, and the variance matrix of estimated error is offered. Through deduction, it can be found out that treated appraise precision can be directly analyzed and deduced without carrying out real measure and reaching estimation value. The simulation result shows that the system performance based on the weighted least square method is much better than the traditional passive location method, and it can be also used for reference to the research of the location algorithm of similar system.

Regulated multi-sited practice for physicians in China: incentives and barriers

The multi-sited practices (MSP) policy in China is expected to exert found implications upon uneven distribution of high-quality medical resources,coupled with the widespread phenomenon of physician's informal practices on multiple sites.Since 2009,the Chinese government has upheld a policy that allows medical doctors to take on secondary jobs for the purpose of encouraging talented physicians to work in medical institutions and underdeveloped areas where their expertise is badly needed.This paper focuses on the MSP policy in China by examining how it has been implemented at the national and provincial levels through a review of government documents,statistical information,academic papers,and newspaper accounts.Despite the MSP policy is to offer incentive to physicians at first-tier public hospitals to work in primary care centers or areas where their expertise,the evidence is hard to find.This paper identifies several outstanding challenges to the MSP policy:(1) The importance of the MSP policy which is made by the central government is perceived differently at the provincial level by health officials and hospital administrators;(2) Interest in implementing the policy varies among physicians and hospitals;(3) Concerns about quality control and legal responsibilities for medical doctors to take on secondary jobs are widespread.China's MSP policy differs from the 'dual practice' policy in many countries because of its top-down approach,with its emphasis on government's role and administrative intervention.This paper's findings may be useful for other developing countries where medical expertise is unevenly distributed due to market forces.On the special occasion of the Alma Ata Declaration's 40th anniversary and for the sake of promoting Universal Health Coverage,this paper also aims at sharing with other countries China's experience in trying to develop an adequate force of primary care professionals.At the end of this paper,a four-point proposal is made for policy makers in China.

基于VMD-CWT和CNN数控机床刀具健康诊断方法的研究

在数控机床切削过程中,刀具的健康状况直接影响产品的加工质量。因此,为了对刀具的健康状态作出诊断评价,本文提出了一种基于变分模态分解(VMD)和连续小波变换(CWT)特征提取与卷积神经网络(CNN)的刀具健康诊断方法。该方法首先采集不同健康状态下的刀具在切削时的振动信号,然后经VMD分解成若干IMF分量,并求解每个IMF分量的相关系数,选取相关系数较大的分量进行信号重构;其次采用连续小波变换来构造重构信号的时频图;最后将得到的时频图输入构建的CNN模型中,通过多层卷积、池化处理得到信号特征与刀具健康状态之间的准确映射,进而实现刀具的健康诊断。经实验验证表明,本文所提方法的识别准确率达到98.9%,具有良好的状态识别能力和泛化性,可为刀具健康诊断方法提供一定的理论依据。

基于LabVIEW2011对心音信号CWT滤波的实现

目的:为除去杂波,同时克服STFT对分辨率的限制,提高心音分析的准确性和实时性。方法:引入CWT滤波的理论,基于该理论构造算法,利用LabVIEW2011编程在应用上进行实现。结果:利用CWT滤波的方法成功对杂波滤除。结论:采用上述方法能够实现对于心音信号优化处理。

基于YOLOX的分布式PCCP断丝自动监测方法研究

为了研究一种针对预应力钢筒混凝土管(PCCP)断丝信号的识别技术,设计了使用DN4000mm的PCCP建成模拟测试环境,在非常接近实际工况的情况下进行了1∶1原型断丝监测试验。试验将采集到的断丝信号与之前采集到的运行管道内的噪声信号相互组合,分别通过连续小波变换(CWT)和同步挤压小波变换(SWT)两种方式转化的时频谱图作为断丝信号数据集,采用基于YOLOX的目标检测算法通过提取数据集中断丝信号时频谱图图像特征来判断断丝事件的发生。两组数据集训练出模型的准确率、召回率、F1分数,误检率均达到100%、100%、1和0%。随后的剪枝计算阶段,在保证检测精度不变的前提下,SWT组的剪枝率高于CWT组,最终模型大小仅为1.36 MByte。可见,通过SWT变换得到的时频谱图可以在不损失精度的前提下更大的简化YOLOX模型,为开发PCCP断丝监测系统的嵌入式部署提供了参考。

基于连续小波变换和模型无关元学习的压燃式活塞发动机气门故障诊断研究

针对压燃式活塞发动机缸盖表面振动信号样本少及传统故障诊断方法特征提取和选择困难的问题,提出了一种基于连续小波变换(continuous wavelet transform,CWT)和模型无关元学习(model agnostic meta learning,MAML)的压燃式活塞发动机气门间隙异常故障诊断方法。通过将CWT的特征提取能力和MAML的快速学习能力相结合搭建故障诊断模型。试验结果表明该方法能有效识别气门间隙故障,并且其准确率高于传统基于CWT和卷积神经网络(convolutional neural network,CNN)的故障诊断方法。通过跨域故障对比试验,研究了不同气门故障类型对模型诊断能力的影响,验证了该方法在解决小样本和跨域故障问题时具有更高的准确率和泛化能力。

基于Log-Polar和DT-CWT的旋转不变纹理分类算法

提出了一种基于对数-极坐标变换(Log-Polar)和双树复数小波变换(DT-CWT)的旋转不变纹理分类算法。该方法首先对纹理图像进行对数-极坐标变换将旋转转化为平移,再用具有平移不变性的双树复数小波对变换后的图像滤波并计算各子带的能量值组成旋转不变特征向量,最后利用支持向量机算法实现纹理图像的分类。将该方法与其它旋转不变纹理分类算法进行比较,实验结果表明,提出的算法能有效地提高正确分类率。