遗传算法优化的模糊神经网络在刀具磨损诊断中的应用

曹伟青; 傅攀; 张尔卿

doi:10.13433/j.cnki.1003-8728.2014.1115

遗传算法优化的模糊神经网络在刀具磨损诊断中的应用

doi: 10.13433/j.cnki.1003-8728.2014.1115

1.
西南交通大学机械工程学院, 成都 610031

基金项目:

中央高校基本科研业务费专项资金项目(SWJTU12CX039)

国家科技支撑计划项目(2009BAG12A01-E04)资助

详细信息

作者简介:
曹伟青1978- ),工程师,研究方向为机械故障诊断及智能监测,caoweiqing@swjtu.cn。

通讯作者:
傅攀,教授,博士,博士生导师,pfu@swjtu.cn。

计量
- 文章访问数: 185
- HTML全文浏览量: 50
- PDF下载量: 5
- 被引次数: 0
出版历程
- 收稿日期: 2013-04-06

Application of GA-fuzzy-neural Networks in Tool Wear Diagnosis

1.
School of Mechanical Engineering, Southwest Jiaotong University, Chengdu 610031

摘要

摘要: 针对刀具磨损监测时,采集的振动信号含有强烈的背景噪声,难以提取故障频率的问题,提出采用形态滤波消噪后进行经验模态分解来提取故障频率;同时,为了准确监测刀具的磨损状态,将提取的故障特征输入到遗传算法优化的模糊神经网络对刀具的磨损进行识别,模糊神经网络的基函数采用B样条基函数.传统的网络学习算法采用梯度下降法,这在学习过程中容易陷入局部最小,论文采用遗传算法寻求全局的最优解.实验表明,该方法能有效地应用于强噪声背景下的刀具磨损识别.
- 经验模态分解 /
- 形态滤波 /
- B样条基函数 /
- 模糊神经网络 /
- 遗传算法
Abstract: In view of the strong background noise involved in the fault signal of tool wears and the difficulty to obtain fault feature frequencies, in this paper, a fault feature extraction method was proposed based on morphological filters and combining with empirical mode decomposition. At the same time,tool wears were identified by genetic algorithm(GA)-fuzzy-neural networks with B-spline membership functions. Fuzzy neural networks are traditionally trained by using gradient-based methods,and may fall into local minimum during the learning process. So,the genetic algorithm was adopted for global optimization in this study. The experimental results show that the diagnosis approach put forward in this paper can effectively identify tool wears in strong background noise.
- B-spline membership function /
- empirical mode decomposition /
- failure analysis /
- feature extraction /
- fuzzy neural netrowks

HTML全文

参考文献(13)

[] 杜必强,王松岭,唐贵基形态学降噪在转子分形故障状态识别中的应用[J]中国电机工程学报,2008,28(26);71-76Du B Q,Wang S L, Tang G J. Application ofmorphological de-nosing in rotor fault state identifying byfractaal method[J]Proceedings of the CSEE,2008,28(26);71-76(in Chinese)
[2]	孙斌,黄胜全,周云龙,等-种基于经验模式分解的气液两相流流型识别方法[J]仪器仪表学报,2008,29 (5):1011-1015Sun B,Huang S Q,Zhou Y L, et al. Identificationmethod of gas liquid two phase flow regime based onempirical mode decomposition[J]Chinese Journal ofScientific: Instrument, 2008,29(5):1011-1015(in Chinese)
[3]	行鸿彦,许瑞庆,王长松基于经验模态分解的脉搏信号特征研究[J]仪器仪表学报,2009,30(3);596-602Xing H Y, Xu R Q,Wang C S. Pulse signal featureresearch based on empirical mode decomposition[J]Chinese Journal of Scientific: Instrument, 2009,30(3):596-602(in Chinese)
[4]	郝如江,卢文秀,褚福磊形态滤波在滚动轴承故障声发射信号处理中的应用[J]清华大学学报,2008,48(5):812-815HaoR J, Lu WX, Chu F L. Morphology filters foranalyzing roller bearing fault using acoustic emissionsignal processing[J]Journal of Tinghua University,2008,48(5):812-815(in Chinese)
[5]	王太勇,王正英,青永刚,等基于SVD降噪的经验模式分解及其工程应用[J]振动与冲击,2005,24(4);96-98Wang T Y, Wang Z Y, Xu Y G, et al. Empirical modedecomposition and its engineering applications based onSVD de-noising[J]Journal of Vibration and Shock,2005,24(4);96-98(in Chinese)Shin K, Hammond J K,White P R. Iterative SVDmethod for noise reductaion of low-dimension chaoticttime series[J]Mechanical Systems and Signalprocessing,1999,13(1):115-124
[6]	汤宝平,蒋永华,张祥春基于形态奇异值分解和经验模态分解的滚动轴承故障特征提取方法[J]机械工程学报,2010,46(5);37-42Tang B P,Jiang Y H,Zhang X C. Feature extractaionmethod of rolling bearing fault based on singular valuedecomposition-morphology filter and empirical mode de-composition[J]Journal of Mechanical Engineering,2010,46(5):37-42, 48(in Chinese)
[7]	桑海峰,何大阔,张大鹏基于支持向量机与遗传算法的发酵过程软测量建模[J]东北大学学报,2007,28(6):781-784Sang H F, He D K, Zhan G P. Soft-sensing modeling ofa fermentation process through support vectaor machinesand genetic: algorithms[J]Journal of NortheasternUnivensity,2007,28(6);781-784(in Chinese)
[8]	李世超,石秀华,崔海英,等基于遗传小波神经网络的双余度电机故障诊断[J]振动、测试与诊断,2009,29(2);223-226Li S C,ShiXH,Cui H Y, et al. Fault diagnosis basedon genetic: algorithms wavelet neural network in dual-redundancy brushless DC motor[J]Journal ofVibration, Measurement&Diagnosis, 2009,29(2):223-226(in Chinese)
[9]	周伽,陈果光谱油样分析数据的神经网络多变量预测技术[J]仪器仪表学报,2006,27(4);389-393Zhou J, Chen G. Artificial neural network multi-variableforecasting model for spectaral oil analysis data[J]Chinese Journal of Scientific: Instrument,2006,27(4):389-393(in Chinese)
[10]	林瑞霖,周平基于EMD和神经网络的气阀机构故障诊断研究[J]海军工程大学学报,2008,20(2);48-51Lin R L, Zhou P. Fault diagnosis of valve train based onEM D and neural network[J]Journal of NavalUniversity of Engineering, 2008,20(2):48-51(in Chinese)
[11]	胡爱军,安连锁,唐贵基HILBERT-HUAN(;变换端点效应处理新方法[J]机械工程学报,2008, 44 (4);154-158Hu A J, An L S, Tang G J. New process method for endelfectas of HILBERT-HUANG transform[J]Journal ofMechanical Engineering, 2008,44(4):154-158(in Chinese)
[12]	杜必强,王松岭,唐贵基转子系统振动信号形态滤波器的设计[J]中国机械工程学报,2008,6(2);243-248Du B Q,Wang S L, Tang G J. Morphological filterdesign for vibration signals from rotor system[J]Chinese Journal of Constructaion Machinery,2008,6(2):243-248(in Chinese)
[13]	Wang WY, Li Y H. Evolutionary learning of BMFfuzzy-neural networks using a reduced-form geneticalgorithm[J]IEEE Trans. Syst,Man, Cybern. B,2003,33:966-976