中国机械工程
中國機械工程
중국궤계공정
CHINA MECHANICAl ENGINEERING
2014年
18期
2473-2477
,共5页
随机共振%遗传算法%信噪比%B样条神经网络
隨機共振%遺傳算法%信譟比%B樣條神經網絡
수궤공진%유전산법%신조비%B양조신경망락
stochastic resonance(SR)%genetic algorithm%signal-to-noise ratio%B-spline neural net-work
针对刀具的早期故障监测中因存在强烈的背景噪声而难以提取故障特征的问题,提出了基于二次采样随机共振消噪和B样条神经网络智能识别的故障诊断方法。首先利用在随机共振过程中,噪声增强振动信号的信噪比特性,将刀具振动信号进行随机共振输出,提取有效特征,再输入到B样条神经网络进行智能识别,进而获得刀具的磨损值。同时,为了得到与输入信号最佳匹配的随机共振参数,提出了基于遗传算法的多参数同步优化的自适应随机共振算法,克服了传统随机共振系统只实现单参数优化的缺点。实验结果表明,该方法能实现弱信号检测,能有效地应用于刀具磨损故障诊断中。
針對刀具的早期故障鑑測中因存在彊烈的揹景譟聲而難以提取故障特徵的問題,提齣瞭基于二次採樣隨機共振消譟和B樣條神經網絡智能識彆的故障診斷方法。首先利用在隨機共振過程中,譟聲增彊振動信號的信譟比特性,將刀具振動信號進行隨機共振輸齣,提取有效特徵,再輸入到B樣條神經網絡進行智能識彆,進而穫得刀具的磨損值。同時,為瞭得到與輸入信號最佳匹配的隨機共振參數,提齣瞭基于遺傳算法的多參數同步優化的自適應隨機共振算法,剋服瞭傳統隨機共振繫統隻實現單參數優化的缺點。實驗結果錶明,該方法能實現弱信號檢測,能有效地應用于刀具磨損故障診斷中。
침대도구적조기고장감측중인존재강렬적배경조성이난이제취고장특정적문제,제출료기우이차채양수궤공진소조화B양조신경망락지능식별적고장진단방법。수선이용재수궤공진과정중,조성증강진동신호적신조비특성,장도구진동신호진행수궤공진수출,제취유효특정,재수입도B양조신경망락진행지능식별,진이획득도구적마손치。동시,위료득도여수입신호최가필배적수궤공진삼수,제출료기우유전산법적다삼수동보우화적자괄응수궤공진산법,극복료전통수궤공진계통지실현단삼수우화적결점。실험결과표명,해방법능실현약신호검측,능유효지응용우도구마손고장진단중。
In view of the difficulties of fault feature extraction from strong background noise in tool wear early fault diagnosis ,a method was proposed based on twice sampling SR and B-spline neural net-work .First ,SR was employed to remove noise in tool wear vibration signals because of its benefits for enhancing the signal-to-noise ratio ,then ,tool wears with the good fault features were identified by B-spline neural network .In order to improve the deficiency of a single parameter be optimized in the tra-ditional SR and achieve the best SR parameters ,an adaptive SR was proposed based on genetic algo-rithm ,which realized multi-parameter synchronous optimization .The experimental results show that this method can realize the weak signal detection and apply to tool fault diagnosis effectively .