用户名: 密码: 验证码:
液力传动元件流场实验及流动图像判读系统研究
详细信息    本馆镜像全文|  推荐本文 |  |   获取CNKI官网全文
摘要
液力传动是以液体为工作介质,利用介质动能传递动力的叶轮传动机械,其基本形式为液力偶合器与液力变矩器。目前,液力传动已广泛地应用于军事工业和各民用工业部门,提高液力传动元件的效率、降低能耗对国民经济的发展具有重要意义。
    与国外相比,国内对液力传动现代设计理论与方法的研究成果并不多,并且,流场测试研究的则更少。针对目前国内研究状况,吉林大学液力传动课题组申请了“液力变矩器现代设计方法”的“国家自然科学基金课题”,这是唯一的液力传动方面研究三维流动的国家自然科学基金课题。本文主要是针对这个课题,研究液力传动元件流动显示方法及数据判读系统。
    为了提高液力传动元件的性能,必须对液力传动元件内部流动进行测试,以增进对其内部流动的了解,掌握内部流场的结构与分布,验证理论的正确性。流场测量有两个方面,即流动显示和流动测量。流动显示的任务是使流体传输的过程可视,而流动测量的任务是获得流体传输的定量信息,它们是相辅相成的。
    粒子成像测速技术集“可视化”与“定量测量”于一体,是一种现代的流动显示与测量技术,是光学测量技术、计算机应用技术、图像处理技术相结合的产物。粒子成像测速技术得到的图像对其粒子位移的大小和方位的判读采用跟踪识别的判读方法。
    粒子图像测速技术分两大部分:粒子图像底片形成技术及粒子图像底片判读技术。所以,本文所涉及的内容总体上分为两部分,一是实验研究,二是流动图像判读系统的研究。
    本文主要是针对实验所得的示踪粒子图片,利用数字图像处理与分析的知
    
    
    识,以及模式识别方面的理论,对流动图像的判读系统进行研究,进而对液力传动元件的流场进行初步的分析和研究。对内部流场进行定性和定量的分析。以前在做数值模拟分析时,初始速度的给定仅凭经验。本文对实验所得的示踪粒子速度的求解,可作为数值模拟的精确的初始条件。
    在实验方面,本文利用“LSFM150型激光切面流场测量系统”获取流动图像。在实验时,充分考虑示踪粒子的跟随性,为了得到清晰的流动图像,采用了三曝光的方法。
    在流动图像的判读方面,本文针对实验的特点,对单帧图像的粒子进行识别和跟踪,变成矢量式。研制出适合本实验流动图像特点的流动图像判读系统,该系统以实验得到的图像为基础,分析图像的特点,利用Visual C#语言,结合图像处理与模式识别的知识研制出一套适合本实验特点的流动图像判读系统。
    利用已做好的流动图像判读系统软件包,对实验所得的流动图像进行识别,并且也判读了液力偶合器的流动图像,结果表明此系统是判读流动图像的
    行之有效的方法。
Hydrodynamic transmission is impeller transmission machine, which utilize kinetic energy of liquid impress motivity. The basic form of hydrodynamic transmission is hydrodynamic torque converter and hydrodynamic coupling. Now , along with hydrodynamic transmission more and more applies in military affairs industry and all kind of civil industry, it has great meaning in the development of the people’s economy to increase the efficiency of hydrodynamic transmission component and decrease its energy loss.
    To be compared with abroad, the study harvest of modern design methods of hydrodynamic transmission is few in domestic, and the research of flow measurement is also few. Aimed at the study status in domestic, the research team of hydrodynamic transmission of Jilin University applied the project ‘The modern design method on the torque convert ‘belonged to the ‘National Natural Science Fund’. This is the only project on the research of three-dimension flow in the fluid drive that was funded by National Natural Science Fund. This paper studies flow visualization method and data determining system of hydrodynamic transmission component.
    In order to improve the performance of hydrodynamic transmission component, we must test the internal flow of hydrodynamic transmission component so that we can understand the internal flow, hold construction and distributing of internal field, and confirm the correctness of design theory. There are two aspects in measurement of flow field. One is flow visualization; the other is measurement of flow. The task of flow visualization is to make the process of liquid transfers visible, and the measurement of flow is to gain the ration information of liquid transfers, and they supplement each other.
    Particle imaging velocity technique includes flow visualization and quantitative measurement. It is a modern technique of flow visualization and measurement of flow, and it also utilizes optics measurement technique and
    
    
    computer technique and image processing technique. It adopts trail and identifying method to determine the displacement and orientation of particle according to the image which was obtained by particle imaging velocity technique.
    There ate two main parts of particle imaging velocity: forming particle image negative technique and determining particle image negative technique. Therefore, the content of this paper also has two parts. One is experiment study; the other is study on the system of determining flow image.
    This dissertation put forward a determining system of flow image aimed at tracer particle image by experiment, using digital image processing and analysis、 pattern recognition and artificial intelligence. Qualitative and quantitative analysis is used on flow field of hydrodynamic transmission component with this system. By this system, it can find out the velocity which can be used as accuracy initial condition of numerical value simulation.
    At the aspect of experimental research, the article makes use of the LSFM150 flow measurement system to obtain flow image. We consider follow character of tracer particle in order to gain clear image it adopts three-exposal method during experiment.
    At the aspect of determining of flow image, the article identifying and tracking the trail of particle of single frame, and translate it into velocity vector. This paper put forward a determining system of flow image which is fit for the character of the flow image obtained by experiment. Based on the image obtained by experiment, this system analyses the character of the image, and uses Visual C# and combine image processing and pattern recognition to study this system.
    At last, using the determining system of flow image to process the flow image which is obtained by experiment, and determining the flow image of hydrodynamic coincidence, it turns out the system is an effective method of processing flow image.
引文
罗邦杰,液力机械传动,北京:人民交通出版社,1983
    李有义,液力传动,哈尔滨工业大学出版社,2000
    范洁川等,流动显示与测量,机械工业出版社,1997
    戴昌晖等,流体流动测量,航空工业出版社,1991
    赵志新,液力变矩器流场数值分析,吉林大学硕士学位论文,2003.3
    李福忠.流动显示学在风机领域中最新进展(之三)—示踪子的跟随性与示踪子法的应用.风机技术.1996(3)风速仪
    张玮、王元、徐忠,叶轮机械内部流动测量及动静相互作用的实验研究进展,流体机械,2001(8)
    方杰,液力变矩器三维流动的研究,吉林大学硕士学位论文,2003.3
    吴根兴,流动显示与拓扑分析,南京航空学院,1986
    马文星等,液力传动元件内部流场测量的激光切面法,农业工程学报,1997(6)144~147
    张军等,三曝光三相关PIV技术研究,实验力学,2001(3)19~25
    施业等,PIV测速中实时可调同步脉冲光照系统的设计与应用,河海大学学报,2002(1)54~57
    盛森芝、徐月亭,近十年来流动测量技术的新发展,力学与实践,2002(5)
    马文星,国外车辆液力传动研究现状及其展望,汽车工程,1996(4)1~4
    Msdn Library For Visual Studio.Net 2003
    李兰友、王学彬,Visual C#图像处理程序设计实例,北京:国防工业出版社,2003.4
    东方人华,Visual C#.NET范例入门与提高,清华大学出版社,2003.9
    
    Mayo,J.著、王启丁等译,C#技术内幕,北京:机械工业出版社,2003.1
    王昊亮、李刚等编,Visual C#程序设计教程,清华大学出版社,2003.10
    吕凤军,数字图像处理编程入门——做一个自己的Photoshop, 北京:清华大学出版社2000.3
    周丽华,用VB实现医学图像的增强和平滑, 中华医药与健康杂志2003 第三期
    黄贤武 王加俊 李家华 编著,数字图像处理与压缩编码技术,成都:电子科技大学出版社,2000.12
    沈庭芝、方子文,数字图像处理及模式识别,北京:北京理工出版社,2001.9
    刘榴梯,实用数字图像处理,北京:北京理工大学出版社,2001.9
    刘文萍 吴立德. 图像分割中阈值选区方法比较研究. 模式识别与人工智能,1997,10(3)
    Nikhil R P,Sankar K P. A Review on Image Segmentation Techniques. Pattern Recognition,1993,26:1277~1294
    张毅军,吴雪前,夏良正. 二维熵图像阈值分割的快速递推算法. 模式识别与人工智能,1997,10(3)
    Abutaleb A S. Automatic thresholding of gray-level picture using two-dimensional entropy. CVGIP,1989,47:23~32
    Chen W T, et al. A fast two –dimensional entropic thresholding algorithm. Pattern Recognition, 1994, 27:885~889
    章毓晋,图像分割, 北京:科学出版社,2001
    崔屹,图像处理与分析——数学形态学方法及应用,北京:科学出版社,2000
    朱安定, 计算机图像处理技术辅助精子运动能力分析研究浙江大学硕士学位论文2002.5
    崔屹,图像处理与分析——数学形态学方法及应用,北京:科学出版社,
    
    
    2000
    谷口庆冶,数字图像处理——基础篇,北京:科学出版社,2001
    谷口庆冶,数字图像处理——应用篇,北京:科学出版社,2001
    Kenneth R. Castleman, Digital Image Processing, Prenticle Hall Inc., a simon & Schuster Company,1996
    Kenneth R. Castleman著,朱志刚等译,数字图像处理,北京:电子工业出版社,2002,2
    吴世法,近代成像技术与图像处理,国防工业出版社,1997.3
    李在铭,数字图像处理压缩与识别技术,电子科技大学出版社,2000,11
    苏光大,微机图象处理系统,清华大学出版社,2000,7
    孙兆林,MATLAB 6.图像处理,清华大学出版社,2002,5
    周培德,平面散乱点线集三角剖分的算法,计算机辅助设计与图形学报,2003(9)
    Adrian R J, Particle-Imaging Techniques for Experimental Fluid Mechanics. Annu Rev Fluid Mech, 1991,21~25
    Godde P, Die stromung in hydrodynam ischen kupplungen bei schcufelg it te rvariation . Dissertation, Ruhr-Universitat bochum,1995.109
    任宏周等,新型图像传感器CCD的原理及应用,中州煤炭,1998(1)10~11
    Marzonk V M, Hart D P. Asymmetric auto-correlation function to resolve directional am biguity in PIV images [J],Experiments in Fluids, 1998,25:401~408
    Xu Jie ,A Study of Image Processing on PIV ,Journal of Ship Mechanics,
    1998,8:1~6
    靳斌,杨冠玲等,一种利用示踪粒子群体运动特征的PTV方法,光学技术,2000(1)
    徐涛,数值计算方法,长春:吉林科学技术出版社,1998
    
    李聪、张勇、高智,一种新的聚类算法,模式识别与人工智能,1999(6)
    勒中鑫,数字图像信息处理,北京:国防工业出版社,2003.1
    Chinese Journal of Computational Mechanics,Vol.19,No.1,February 2002,pp20~25。

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700