首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献,搜索用时 156 毫秒
1.
为精确捕捉诱导轮中空化的行为特征并揭示其非定常流动机理,采用大涡模拟(LES)对一凝水泵诱导轮在设计工况、临界空化数下单独进行数值分析。基于Omega涡识别方法,对诱导轮内空化流动的流动结构及空化泡的关系进行研究。结果表明,Omega涡识别技术可以很好反映流动结构(剪切流与旋转流)与空化泡的关系:设计流量、临界空化数下,诱导轮叶片入口附近空化泡主要集中在轮缘附近的回流区内,回流区内部流动主要以剪切流的形式存在;而在流道中的回流区内流动主要以旋转运动为主,且空化泡主要集中在回流区与主流的剪切层区。  相似文献   

2.
采用基于OpenPose的单人姿态估计算法以及多人姿态检测算法,将摄像头采集到的人的睡眠图像进行处理,实现对人脸的关键点的定位,人手的关键点的定位以及人体身体的各个关节点的定位;采用基于人体关键点提取的睡姿识别算法来判断人体的睡眠姿态——左侧卧、右侧卧、仰卧以及俯卧.实验结果表明本文采用的方法能够较好地适应复杂的背景,并且具有较高的识别准确率以及运算速率,识别准确率可达92.5%,而运算速率可达每秒9.8帧.  相似文献   

3.
紫檀属中的木材有很多属于名贵木材,不同树种之间十分相似。传统的木材识别方法多以木材解剖学为主,通过观察木材的切片结构特征对木材的树种进行判断,这类方法虽有较高的识别精度,但是其识别工艺较为复杂而且技术难度也相对较高。与木材解剖学相对应的是利用图像信息或光谱信息的木材树种识别方法,该类方法虽具有较为简单的识别工艺,但是在对同属相似木材树种进行识别时,往往不能够取得较好的识别效果。提出了一种基于木材切面光谱特征和纹理特征相融合的木材树种识别方法,该方法不仅识别工艺简单、自动化程度高,而且具有较高的识别精度。首先通过数码相机和光谱仪采集木材切面的图像信息和光谱信息,然后分别使用纹理特征提取方法和光谱特征提取方法提取两类特征的特征向量,接下来使用基于典型相关分析的特征级融合方法将这两个特征向量进行融合,最后使用支持向量机对融合后的特征向量进行分类识别。为了验证方法的有效性,以市场中常见的5种紫檀属树种的三个切面为研究对象,对这些木材树种进行了识别。实验结果显示,单独使用纹理特征的识别正确率最高为80.00%,单独使用光谱特征的识别正确率最高为94.40%,使用融合的特征最高的识别正确率可达99.20%。还将这5种木材树种与其他30种木材树种进行了混合,混合后的木材样本数量可达1750。实验进一步显示,该方法可以对包含紫檀属在内的35种木材的树种进行识别,其正确率可达98.29%。综上所述,木材的纹理特征和木材的光谱特征可以有效的相互补充,从而进一步提高识别正确率。最后还用所提出的方法与目前主流的方法进行了比较,结果发现所述的木材树种识别方法高于目前主流方法。  相似文献   

4.
针对相位敏感光时域反射系统中传统人工特征提取和模式识别方法实时性差、准确率低的问题,提出一种波网络模式识别方法。该方法通过因果空洞卷积结构充分分析光纤振动信号的时序因果性,通过残差块结构使模型更快收敛,以实现更高的识别准确率和效率。实验结果表明,在对手拍、脚踩、棒击三种信号识别时,与一维卷积神经网络结构和长短期记忆网络结构方法相比,该方法识别准确率高达99.85%;且训练耗时最少,低至96 s,测试耗时也仅为30 ms,满足应用实时性的要求。该模式识别方法既具有高准确率又具有高实时性,对于φ-OTDR系统在周界安防中的应用推广具有重要意义。  相似文献   

5.
孟庆芳  陈珊珊  陈月辉  冯志全 《物理学报》2014,63(5):50506-050506
癫痫脑电信号的自动检测对癫痫的临床诊断与治疗具有重要意义.基于递归图(recurrence plot)的递归量化分析(recurrence quantification analysis,RQA)重现了非线性时间序列的动力学行为,分析了其递归特性,本文提出了基于RQA的癫痫脑电信号特征提取方法.实验结果表明:直接基于RQA特征的癫痫脑电的检测准确率较高,其中直接基于确定率DET的分类准确率可达到90.25%.本文还把提取的RQA特征值和变化系数、波动指数相结合组成特征向量,输入到SVM分类器,实现癫痫脑电信号的自动检测;实验结果表明:该方法的分类准确率可达到99%.  相似文献   

6.
针对身份识别容易被仿冒和造假的问题,提出了一种利用近红外相机捕获手背静脉同时具有活体检测功能的身份识别方法,手背静脉图像提供静脉特征作为身份识别的依据,与此同时获取的脉搏波的周期性特征作为活体检测的标志。利用自行搭建的手背静脉和脉搏波捕获实验装置,研究了70个个体的手背静脉图像以及活体和假体的静脉图像特征,并提出了提高身份识别准确率的算法。采用主成分分析对活体静脉特征向量进行降维,降低分类算法的复杂度,结合马氏距离去除异常样本,以提高识别精度,再采用参数优化的随机森林算法和支持向量机算法实现了手背静脉的精准识别。结果表明:基于手背静脉特征结合随机森林算法和支持向量机算法可以对不同个体进行身份识别,识别准确率分别为99.28%和99.86%,识别时间分别为0.368 s和0.110 s。  相似文献   

7.
对一台凝水泵进行压力脉动实验研究。该凝水泵的叶轮由5长5短叶片组成,叶轮上下游分别设置导流栅和径向导叶。结果表明,该泵的临界汽蚀余量小于1m。低转速条件下,泵的总体压力脉动较高转速条件下低,但同流量下,泵的空化性能较高转速时差。随着泵进口压力的降低,叶频和1/2叶频对应的压力脉动幅值呈现下降趋势,并于空化临界点附近剧增。而小流量工况下,随着空化的发展,叶频被抑制,同时1/2叶频对应的压力脉动幅值呈现不规则变化。  相似文献   

8.
针对中国农田存在种植景观破碎化和复杂的种植结构这一现状,如何实现目标作物的高精度识别与制图对作物产量估算、粮食政策调整和国家粮食安全保障具有十分重要意义。基于Google Earth Engine(GEE)遥感数据处理云平台,提出一种冬小麦不同生育期的种植结构提取方法,该方法以2021年覆盖目标作物关键生育期的多时相Sentinel-2影像为数据源,综合考虑光谱波段特征、光谱指数特征、纹理特征和地形特征等多维特征变量,利用GBDT(gradient boosting decision tree)分类器对不同生育期田块尺度的冬小麦种植面积和空间分布信息进行快速精准提取,并探讨了冬小麦识别的最佳生育期。此外,对比分析了常见的不同分类模型在田块尺度条件下的作物识别性能。以河南陈固镇为研究区开展实验,实验结果显示,冬小麦在起身拔节期的地物识别准确率相对较高,总体分类准确率为94.61%, Kappa系数为92.68%;在抽穗扬花期的识别精度最高,总体分类准确率为97.01%, Kappa系数为95.52%;但在灌浆乳熟期的分类精度偏低,总体分类准确率为86.23%, Kappa系数为81.33...  相似文献   

9.
车型自动识别是智能交通系统的重要组成部分。针对现有车型识别存在的问题,提出利用经验模态分解和支持向量机的车型声频识别方法。将车辆行驶的声音信号进行分解,以分解不同模态的能量作为特征向量,并以此作为训练样本对支持向量机构成的车型识别器进行训练,通过对小汽车和卡车的声音信号处理结果表明:利用车辆声音信号能够正确识别不同的车型,识别准确率达95%,是车型识别的有效方法。  相似文献   

10.
为了提高人体异常行为识别的准确率,采用了一种将多特征结合的异常行为识别算法,主要包括对步行、快跑、慢跑、拳击、双手挥舞、鼓掌六种异常行为进行识别。首先从视频流中提取出人体轮廓,然后从所得的轮廓中提取Hu矩特征与纹理特征。最后通过模板匹配的方法,采用马氏距离度量所需识别的当前行为特征向量与标准模板行为的特征向量之间的相似性,并通过设置相应的阈值判定该行为所属类别。实验证明,该方法比提取单一特征的方法识别率高,且具有一定的实用价值。  相似文献   

11.
Safe operation of kinetic pumps, as liquid movers, can be threatened by cavitation phenomenon in, amongst others. Cavitation is the Achilles’ heel of kinetic pumps. It can cause deterioration of the hydraulic performance, damage of the pump by pitting and material erosion, and structure vibration and noise. Cavitation can appear within the entire range of operating conditions, therefore it must by all means be prevented. To prevent cavitation in a pump we have to know the beginning and development of the cavitation in the pump. For this purpose, the emitted noise in the audible range can be used, amongst other possibilities. Experiments have shown that there is a discrete frequency tone within the audible noise spectra, which is in strong correlation with development of the cavitation process in the pump. Therefore, the discrete frequency tone can be separated from the noise spectra of a cavitating pump and used to detect the incipient of cavitation and its development as well as to prevent the onset of the cavitation process in the pump, by means of initiating an alarm, shutdown, or control signal via an electrical control system.  相似文献   

12.
Cavitation within a pump causes structural vibration with noise, among other by-products. Experiments have shown that there is a discrete frequency or broadband peak within the audible noise spectra, which is in strong correlation with the development of the cavitation process in the pump. Furthermore, the peak of the discrete frequency, or broadband peak, coincides with the net positive suction head (NPSH) critical value, which corresponds to a 3% drop in the total delivery head. Therefore, the discrete frequency tone can be used to detect the incipience of cavitation and its development as well as to determine the NPSH required or critical value with in situ operation of a pump. In this study, we wanted to clarify the mechanism of noise generation, which is responsible for the discrete frequency component and on which cavitation has an important effect. For this purpose, three different measurement methods were used: the first is based on measurement of the sound pressure level (SPL) in the surrounding air, the second is based on measurement of the underwater acoustics and the third is based on measurement of the structural vibration. Experiments have shown that the characteristic discrete frequency tone, which is in close correlation with the cavitation process, is a result of structural vibrations (modes) or resonances caused by implosion of bubbles and bombardment of the inner surfaces of the pump.  相似文献   

13.
As a complex field-circuit coupling system comprised of electric, magnetic and thermal machines, the permanent magnet synchronous motor of the electric vehicle has various operating conditions and complicated condition environment. There are various forms of failure, and the signs of failure are crossed or overlapped. Randomness, secondary, concurrency and communication characteristics make it difficult to diagnose faults. Meanwhile, the common intelligent diagnosis methods have low accuracy, poor generalization ability and difficulty in processing high-dimensional data. This paper proposes a method of fault feature extraction for motor based on the principle of stacked denoising autoencoder (SDAE) combined with the support vector machine (SVM) classifier. First, the motor signals collected from the experiment were processed, and the input data were randomly damaged by adding noise. Furthermore, according to the experimental results, the network structure of stacked denoising autoencoder was constructed, the optimal learning rate, noise reduction coefficient and the other network parameters were set. Finally, the trained network was used to verify the test samples. Compared with the traditional fault extraction method and single autoencoder method, this method has the advantages of better accuracy, strong generalization ability and easy-to-deal-with high-dimensional data features.  相似文献   

14.
李刚  贺昱曜 《光子学报》2014,39(8):1405-1408
针对受光照不均影响的路面裂缝图像,提出一种基于Sobel算子和最大熵法的图像分割算法,并采用长线段与原图进行与操作和判断黑色像素所占比例的方法去除图像孤立噪声点.根据不同类型裂缝的几何形态,提取投影向量、分布密度和空洞数等特征值作为路面裂缝分类的依据,设计径向基函数神经网络的分类器实现对裂缝的准确分类.实验结果表明,较传统全局阈值算法,本文算法对光照不均图像的处理不仅能很好的提取裂缝边缘,且具有很强的抗噪能力,对路面裂缝的分类准确率高.  相似文献   

15.
Cavitation is a common phenomenon in a fluid circuit especially wherever local pressure is lower than fluid saturated pressure. The cavitation negatively affects a fluid system and structure in different ways: i.e. erosion, flow rate reduction, noise and vibration. In order to diminish cavitation, adding some nanomaterials seem to be applicable in different ways. This research aims at assessing the effects of CMC (Carboxy methyl cellulose) additives on incipient cavitation by analyzing the resultant change in low frequency acceleration. Furthermore, this study attempts to examine the accuracy of low frequency acceleration for detecting incipient cavitation.  相似文献   

16.
Propeller sheet cavitation is the main contributor to high level of noise and vibration in the after body of a ship. Full measurement of the cavitation-induced hull pressure over the entire surface of the affected area is desired but not practical. Therefore, using a few measurements on the outer hull above the propeller in a cavitation tunnel, empirical or semi-empirical techniques based on physical model have been used to predict the hull-induced pressure (or hull-induced force). In this paper, with the analytic source model for sheet cavitation, a multi-parameter inversion scheme to find the positions of noise sources and their strengths is suggested. The inversion is posed as a nonlinear optimization problem, which is solved by the optimization algorithm based on the adaptive simplex simulated annealing algorithm. Then, the resulting hull pressure can be modeled with boundary element method from the inverted cavitation noise sources. The suggested approach is applied to the hull pressure data measured in a cavitation tunnel of the Samsung Heavy Industry. Two monopole sources are adequate to model the propeller sheet cavitation noise. The inverted source information is reasonable with the cavitation dynamics of the propeller and the modeled hull pressure shows good agreement with cavitation tunnel experimental data.  相似文献   

17.
空化流动诱导离心泵低频振动的实验研究   总被引:2,自引:0,他引:2  
空化流动严重影响泵安全稳定运行,为充分认识泵内空化发展程度及其诱导的低频振动特性,设计了一台离心式模型泵作为研究对象,同时采用高速摄影及振动加速度测试手段,实现了叶轮内部空化流动的可视化及对应的泵体低频振动信号提取,分析了空化泡形态随空化发展的演化规律,对比了空化发生前后泵体低频振动频谱特性,探讨了泵体总振级水平、离散频率下振动加速度幅值随空化发展程度的变化,提出振动临界汽蚀余量可作为空化程度的另一判据,叶频时的振动加速度幅值变化亦可作为泵内空化程度的表征。  相似文献   

18.
This paper describes an analytical method for estimating the coolant pump-induced acoustic pressure distribution in the inlet annulus of a pressurized water reactor. To simplify the problem, flow velocity is ignored and the walls bounding the annular cavity are assumed to be rigid. The phenomenon of beating due to slight differences in the pump blade passing frequencies is included in this analysis. An experiment based on a simple laboratory model is carried out to verify the theory, small loudspeakers being used to simulate the pumps and air as the fluid medium. Agreement between theory and measurement is good.  相似文献   

19.
空化检测对于保障离心泵运行的安全性和可靠性具有重要意义,已有研究侧重于信号采集和特征提取,对于空化诱发的振动噪声形成机理研究不够深入。为了实现离心泵空化状态的准确表征和有效识别,本文建立了基于信号调制理论的流体机械振动噪声信号模型,将流体激振信号和调制信号视为空化表征的有效信息成分,在此基础上提出了一种基于频带能量和峭度的主导频带时频分析方法,并结合卷积神经网络实现空化状态智能识别.最后,仿真信号和实际数据的分析结果验证了流体机械信号模型的合理性,也证明了所提出的主导频带时频分析方法的有效性。  相似文献   

20.
离心油泵输送粘性流体空蚀性能实验研究   总被引:1,自引:0,他引:1  
本文通过实验研究了离心油泵输送粘性流体时的空蚀性能和油泵的空蚀性能与输送介质粘度的关系,研究了油泵的性能参数(流量、扬程、功率和效率)随进口真空度的变化规律,并与输送清水时的空蚀性能进行了对比,研究了进口真空度对油泵水力性能的影响。离心油泵的空蚀性能随输送介质的不同存在很大差异。实验结果表明:在相同的流量工况下,输送粘性流体比输送清水具有较高的临界空蚀余量。  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号