摘要
数控机床是数控技术在工业生产中的典型应用,数控机床的“智能化”是数控技术的发展方向。数控机床状态监测与故障诊断技术是数控机床“智能化”的重要内容。该方向的研究不仅具有一定的理论研究意义,而且具有明确的应用背景。
本文在综合国内、外状态监测与故障诊断技术的基础上,以数控线切割机床为对象,对数控机床的运行特性进行了研究,设计出了数控线切割机床智能状态监测系统。本系统以计算机为主体,通过各种传感器实时采集数控线切割机床运行过程中的特征参数,并对特征参数进行实时分析与处理,利用人工神经网络对数控线切割机床电极丝的工作状态进行了辩识,提出了运行状态综合劣化度的概念,对数控线切割机床运行状态进行了有效的评价。
本系统集数据采集、信号处理、状态监测于一体,是数控线切割机床状态监测与故障诊断系统的重要组成部分。在SCX-Ⅱ型数控线切割机床上进行了电极丝工作状态识别及数控线切割机床运行状态综合劣化度的评价分析实验,实验结果表明,该系统能实时地采集数控线切割机床运行过程中的特征信号,能正确地识别出电极丝的工作状态,根据采集的特征信号能客观地评价数控线切割机床的运行状态,从而验证了所设计的数控线切割机床智能状态监控系统运行的有效性和可行性。
Numeral control machine tool is the representative application of numeral control technology in industrial manufacturing; the intelligence of the numeral control machine tool is the evolutional orientation of numeral control technology. Numeral control machine tool condition monitoring and fault diagnosis system is the important part of the intelligence of numeral control machine, so this paper has not only the theoretic significance but also explicit applying background.
A lot of natural and international condition monitoring and fault diagnosis technology are studied, the running rule of the numeral control machine tool is studied based on wire electric discharge machine (WEDM), a intelligent condition monitoring system for WEDM is designed. In this system, computer is the principal part, representative parameters of the running WEDM were acquired at real-time by sensors, they were filtered and analyzed at real-time, the working state of the WEDM electrode wire was also recognized by artificial neural networks (ANN), the concept of running compositive condition severity factor was bring forward in this paper, the running state of the WEDM was evaluated effectively.
Data acquisition, signal processing, condition monitoring are concentrated in this system, it is the main part of the condition monitoring and fault diagnosis system for WEDM. Experiments of WEDM electrode wire working state recognizing and WEDM running compositive condition severity factor evaluating have been carried out on SCX- II WEDM, the experimental results indicated that this system could acquire the representative parameters of the running WEDM in real-time, could recognize the working state of the electrode wire correctly, could evaluate the running state with the representative parameters objectively, it is also proved the validity and feasibility of the intelligent condition monitoring system.
引文
[1] 虞和济,陈长征,张省,周建男著.基于神经网络的智能诊断.北京:冶金工业出版社,2000:1-10页
[2] 张安华著.机电设备的状态监控与故障诊断技术.西安:西北工业大学出版社,1995:1-8页
[3] 陈进著.机械设备振动监测与故障诊断.上海:上海交通大学出版社,1999:41-51页
[4] 温熙森,胡莺庆,邱静著.模式识别与状态监控,长沙:国防科技大学出版社,1997:43-59页
[5] 张瑞林著.机械故障诊断技术发展现状及展望.第二届全国机械设备故障诊断学术会议论文集,1988
[6] 刘晋春,陆纪培著.特种加工.北京:机械工业出版社,1989:47-55页
[7] 余承业著.特种加工新技术.北京:国防工业出版社,1995:105-167页
[8] 钟秉林,黄仁著.机械故障诊断学.北京:机械工业出版社,1998:80-95,113-187页
[9] 徐章遂,房立清等著.故障信息诊断原理及应用.北京:国防工业出版社,2000:104-185页
[10] 程乾生著.信号数字处理的数学原理.1996:721-730页
[11] 焦李成著.神经网络系统理论.西安:西安电子科技大学出版社.1996:26-50页
[12] 李士勇著.模糊控制·神经控制和智能控制论.哈尔滨:哈尔滨工业大学出版社,1998:91-101页
[13] [美]Abhijit S.Pandya,Robert.B.Macy著.徐勇,荆涛等译.神经网络模式识别及实现.电子工业出版社版社,1999:57-112页
[14] 徐秉铮,张百灵等著.神经网络理论与应用.广州:华南理工大学出版社,1994:249-303页
[15] 袁曾任著.人工神经网络及其应用.北京:清华大学出版社,1999:49-273页
[16] 徐丽娜著.神经网络控制.哈尔滨:哈尔滨工业大学出版社,1999:4-17页
[17] 王旭,王宏等著.人工神经元网络原理及应用.沈阳:东北大学出版社,2000:19-37页,51-63页
[18] 边肇祺,张学工等著.模式识别.北京:清华大学出版社,2000:250-272页
[19] 王永骥,涂健著.神经元网络控制.北京:机械工业出版社,1999:22-67页
[20] 李人厚著.智能控制理论和方法.西安:西安电子科技大学出版社.1999:102-125页
[21] 程惠涛著.基于神经网络的故障预报技术及其在航天器中的应用.哈尔滨工业大学博士论文.2000
[22] 李勇著.人工神经网络及其在汽轮机故障预测与诊断中的应用.东南大学博士论文.1997
[23] 冯恩波,萧德云,方崇智.一种基于时间序列预报神经网络的故障预报方法及其应用.首都高校第二届自动化控制学术论文集.1992
[24] 杨尔辅著.YF75发动机状态监控与故障诊断工程应用系统的研究.北京航空航天大学硕士论文.1996
[25] 唐英著.基于神经网络的机械加工状态趋势分析技术研究.北京科技大学博士论文.1995
[26] [美]David J.King linski Soct Winge George Shepherd著希望图书创作室译.Visual C++6.0技术内幕.北京:北京希望电子出版社,1994
[27] 李博轩等著.Visual C++图形用户界面开发指南.北京:清华大学出版社,2000
[28] 乔林,杨志刚著.Visual C++6.0高级编程技术.中国铁道出版社,2000
[29] 蒋东翔,倪维斗.大型汽轮发电机组混合智能诊断方法的研究.清华大学学报,1999(5):74-78页
[30] 吕俊涛,王永泓.一种燃气轮机状态监测和故障诊断专家系统的研究.船舶工程.1998(3):38-41页
[31] 章伟,罗干英等.电火花线切割断丝特征及断丝预报的研究.华南理工大学学报,1998(9):60-62页
[32] 韩西京,张锦萍.智能诊断中的状态监测与趋势预报.数据采集与处理.1999(1):115-118页
[33] 罗元丰,赵万生等.基于神经网络的电火花加工过程的分类.电加工与模具,2001(1):14-16页
[34] 徐小力,梁福平等.旋转机械状态监测及预报技术的发展与研究.北京机械工业学院学报.1999(4):34-38页
[35] 杨伸记,廖明夫,赵旭民.旋转机械状态监测与故障诊断系统.测控技术,2000(1):22-24页
[36] 温熙森,秦国军.大型复杂机电设备监测诊断体系结构研究.中国机械工程,1997(2):99-101页
[37] 皇甫冉,陈维斌.通用BP神经网络训练系统的实现.微计算机应用.2000(6):321-324页
[38] 王文武,王诚,郝燕玲等.多媒体定时器的定制和使用方法.计算机应用,2000(3):39-41页
[39] Jozef KORBICZ. Neural Networks and Their Application in Fault Detection and Diagnosis. Fault Detection, Supervision and Safety for Technical Progresses, 1997:367-372P
[40] Jay Lee, John Tsai. On-line Fault Detection Using Integrated Neural Networks. Applications of Artifical Neural Networks Ⅲ,1992:436-443P
[41] W. Hu A. G. Starr. A Systematic Arrroach to Integrated Fault Diagnosis of Flexbile Manufacturing Systems. International Journal of Machine Tools & manufacture, 2000:1587-1602P
[42] Marcantonio Catelani, Ada Fort. Fault Diagnisis of Electronic Analog Circuits Using a Radial Basis Function Network Classifier.Measurement, 2000: 147-158P
[43] R.C.M. Yam, P.W. Tse, L. Li. Intelligent Predictive Decision Support System for Condition-Based Maintenance. Advanced Manufacturing Technology, 2001:83-391P
[44] S. Simani, C. Fantuzzi. Fault Diagnosis in Power Plant Using Neural Networks. Information Sciences, 2000:125-136P
[45] Ouali M.-S, Ait-Kadi D.,Rezg N. Fault Diagnosis Model Based on Petri Net with Fuzzy Colors. Computer & Industrial Engineering,1999:173-176P
[46] W. Hu, A.G. Starr, A.Y.T. Leung. Two Diannosis Models for PLC Controlled Flexible Manufacturing Systems. International Journal of Machine Tools & Manufacutre, 1999:1979-1991P
[47] PAUL. R. DRAKE, DIFU PAN. Mutiple Fault Diagnosis for a Machine Tool's Flood Coolant System Using a Neural Network. International Journal of Machine Tools & manufacture, 1996:1247-1251P
[48] D. Yu. Fault Diagnosis for a Hydraulic Drive System Using a Parameter-Estimation Method. Control Eng. Practice, 1997:1283-1291P