共查询到20条相似文献,搜索用时 140 毫秒
1.
目前,动力刀架振动测试方法的缺失及不完善严重制约了数控车床及车铣复合加工中心行业的发展。鉴于此,开发了基于虚拟仪器平台LabVIEW的三向振动检测系统。系统以三轴加速度传感器为传感器件检测刀架振动信息,采集到的信号经信号调理仪放大和调整,提高系统抗干扰性;通过对采集的振动信号进行频谱变换和频率信息的分析,可探究刀架振动原因进而采取相应措施有效减小振动;最后采集到的数据利用报表生成功能进行存储。 相似文献
2.
齿轮的故障特征信息常常隐藏在齿轮的振动信号中,而振动信号则往往通过加速度传感器采集获得。在不同位置所采集到的振动信号,其噪声干扰也往往存在着强弱差异,因此找到传感器对信号采集的敏感位置在信号采集中显得尤为重要。通过采用故障特征参数值分析方法中的不同的故障状态指标作为参考依据,同时在去噪方法中采用能明显提高信噪比的时域同步平均方法进行去噪,结果发现:安装在靠近轴承座附近位置的A04号传感器所采集的振动信号噪声干扰较小;通过加速度传感器采集齿轮箱的振动故障信号时,最敏感位置应存在于轴承座附近。 相似文献
3.
某发动机在服役过程中,有7片I级涡轮叶片连续发生断裂或开裂。本文对断裂的叶片进行了断口宏微观观察、化学成分分析、金相组织检查、性能试验以及叶片排气边R检查。结果表明,叶片的断裂性质为高周疲劳断裂。断裂叶片的化学成分和力学性能符合技术条件的要求;叶片的疲劳源区未发现夹杂等冶金缺陷。7片叶片的断裂位置均在距离榫头底部62mm-67mm处,该位置是四阶振动的最大应力点,叶片的断裂与四阶振动有关。 相似文献
4.
5.
6.
7.
8.
本文研究了电磁冲击作用对TC11钛合金叶片振动疲劳极限的影响,对电磁冲击前后钛合金叶片进行了残余应力测试和显微硬度测试,利用光学显微镜(OM)和扫描电子显微镜(SEM)对叶片材料显微组织进行了表征,并结合有限元分析,探讨了电磁冲击对TC11钛合金叶片疲劳性能影响机理。结果表明,当电流密度为270 A/mm2时,经电磁冲击处理后叶片振动疲劳极限提升幅度约为36.6%,同时叶片失效裂纹深度及位置发生改变。电磁冲击处理过程中叶片整体温升极小,电极与试样接触处最大瞬时温度仅约60℃。叶片显微组织未见明显变化,叶片表层残余压应力及叶片内部平均显微硬度随电流密度的增大而增大,叶片内部显微硬度分布发生改变。分析认为,电磁脉冲能量能够对叶片材料产生微振动效应,调整叶片整体残余应力分布,修复微观损伤缺陷,达到提升叶片疲劳性能的目的。 相似文献
9.
研究了汽轮机叶片振动测试的数据采集方法,采用光栅光纤及电阻应变片分别测量手段进行叶片振动实验。模拟汽轮机运行状态进行了转子运行实验,提出了实验方案,通过实验测试采集了振动信号数据集。应用Apriori算法对数据集进行分析,使用WEKA平台进行数据处理,定位了转子在这一状态下的易损部位,并发现了与叶片振动相关的关联规则。研究表明,应用Apriori算法可有效发现汽轮机叶片振动问题,将其应用于更多工况下汽轮机运行振动优化具有一定指导作用。 相似文献
10.
为解决单一传感器信号易受干扰且能提取的退化信息有限,导致轴承剩余寿命预测精度低的问题,提出一种基于双通道信息融合与门控单元(GRU)神经网络的轴承剩余寿命预测方法。进行轴承寿命试验时,在振动传感器采集信号的基础上增加声发射传感器,弥补单一信号易受干扰的缺点;使用卷积神经网络自动挖掘出包含轴承退化信息的特征,避免传统算法过分依赖专家判断的弊端;通过归一化处理对信息进行融合;最后使用这些数据训练GRU神经网络,利用训练好的门控单元神经网络预测高铁牵引电机轴承的剩余寿命。结果表明:相比单通道数据,双通道数据训练出的门控神经网络模型的预测结果更为准确;门控单元神经网络相比长短时记忆神经网络有更高的轴承寿命预测精确度。 相似文献
11.
Spot welding is affected by many factors and is difficult to monitor the nugget information with single sensor. In this paper, a new monitor system based on sensors array technique was developed for spot welding quality detection. The key part in the system is cross magnetic sensor array. It is composed of six magnetic sensors. An improved algorithm was proposed to extract the signal eigenvalue, which was based on principle component analysis. The results show that in the case of the 60 mm experiment, cross magnetic sensor array works well. And when the eigenvalues range of the cross magnetic sensor array is 0.090 1-0.098 2, the spots quality is good. The analysis of the eigenvalues of the cross magnetic sensor array allows us to determine whether the spot quality is good or bad. 相似文献
12.
Multisensor data fusion in dimensional metrology 总被引:2,自引:0,他引:2
Multisensor data fusion in dimensional metrology is used in order to get holistic, more accurate and reliable information about a workpiece based on several or multiple measurement values from one or more sensors. The theoretical background originates in classical mathematics and statistics, in methods of artificial intelligence (AI) and in the Bayesian fusion approach. Sensor technologies and sensor characteristics influence the data fusion process and determine the gain of information compared to the application of a single sensor. Homogeneous and inhomogeneous sensor configurations lead to complementary, competitive and cooperative information integration with specific advantages depending on the application. The scope includes image fusion, tactile and optical coordinate metrology, coherent and incoherent optical measuring techniques, computed tomography as well as scanning probe microscopes. 相似文献
13.
对风电叶片进行全尺寸结构试验时需要测量其挠度变形,目前使用拉绳传感器测量,但拉绳传感器只能实现单一方向测量,测量误差大且维数单一。为了解决上述问题,推导一种用于叶片三维挠度精准测量的数学模型,进一步结合超宽带无线测距技术搭建了无线测量系统,使用该系统进行叶片静力加载挠度测量试验,并与激光跟踪仪测量结果进行对比。结果表明:该技术可以获得较为准确的挠度值,测量的X、Y、Z三个方向的叶片变形量的均方根误差分别为12.77、15.91、19.87 mm,符合测试要求。 相似文献
14.
15.
针对滚动轴承寿命状态识别过程中,单一传感器蕴含的信息不能全面反映寿命状态的问题,文章提出了一种基于信息融合的滚动轴承寿命状态识别方法。该方法首先采用多路卷积层提取不同传感器的数据特征信息,克服单一信息源的局限性;然后采用多层卷积、池化交替级联的方式,实现多源信息的特征值深度融合,最后采用全连接和多分类函数,实现动轴承的寿命状态识别。通过不同方法的对比实验,结果表明了所提方法能够提高滚动轴承寿命状态识别率,具有较好的可行性。 相似文献
16.
With the development of faster and more sophisticated computer hardware, sensor data can be processed at increasing rates. The increased demand on sensors which results can lead to problems with the validity of sensor information. One such effect that can occur with ultrasonic sensors is the reception of second-time-around echoes. These are most likely to be present when observing a target in front of a large reflector. A method used in radar processing to eliminate such interference is described and its application to an ultrasonics phased array object location system is presented. 相似文献
17.
声发射检测中的压电换能器 总被引:3,自引:0,他引:3
在声发射检测中,传感器是获取信息的关键部件之一。目前绝大部分声发射传感器都是压电换能器。就采用压电原理的声发射传感器的发展及其校准问题作概述,对声发射传感器中的特殊问题孔径效应及其对传感器设计、性能和校准的影响进行讨论。 相似文献
18.
19.
由于单一传感器存在获取信息量有限、抗干扰能力较弱等问题及传统网络模型诊断时间长、诊断率低等现象,采用振动、噪声等多个传感器监测铣刀的磨损状态。提出将深度学习和多传感器相结合的铣刀磨损状态信号监测方法;将经核主元筛选和未筛选的数据分别输入到BP神经网络、RBF神经网络和深度卷积神经网络中进行模式识别,并对识别结果进行对比和分析。结果表明:深度学习和多传感器相结合的铣刀磨损状态监测方法在特征量比较大、数据量比较多的情况下诊断速度、准确率均比较高,在铣刀磨损状态监测中具有明显的优势。 相似文献
20.
E. E. Zasimchuk Yu. G. Gordienko R. G. Gontareva I. K. Zasimchuk 《Journal of Materials Engineering and Performance》2003,12(1):68-76
This article describes the surface relief formation on Al single-crystal plates (sensors) with orientation {100}〈001〉 that
were rigidly attached to weld specimens of aircraft alloy 2024 T351 during fatigue loading. Taking into account the nonuniform
structure of the weld alloy, we located sensors in different zones of the weld specimens to receive information about the
sites of strain localization and probability of destruction. The qualitative and quantitative information about the sensor
relief was extracted by the online method automated on the basis of a stereomicroscope with a charge-coupled device (CCD)
camera attached to a personal computer by a video adapter and frame-grabber card. The quantitative and qualitative relief
parameters are shown to correlate with the deformation prehistory of the weld specimens. Panoramic views were created and
used for fractal analysis of deformation relief. Spatial distributions of information fractal dimension on the sensor surface
for different numbers of cycles were plotted in the shape of contour lines with equal values (equidimensional maps). Equidimensional
maps for sufficiently large sensors allow us to find strain localization sites in specimens and monitor their evolution in
an online regimen. 相似文献