謝秀嫻，付攀，曹偉青 （西南交通大學(xué)，四川 成都 610031）   摘要：隨著現(xiàn)代加工工業(yè)的發(fā)展，對(duì)刀具磨損的監(jiān)測(cè)在保障生產(chǎn)安全和產(chǎn)品質(zhì)量中發(fā)揮著越來越重要的作用。聲發(fā)射技術(shù)是刀具磨損監(jiān)測(cè)的一種新方法。在車削加工過程中采集聲發(fā)射信號(hào)，用聲發(fā)射信號(hào)對(duì)刀具磨損狀態(tài)進(jìn)行識(shí)別。利用小波包分解技術(shù)對(duì)信號(hào)進(jìn)行分析，得到有效的特征量作為BP神經(jīng)網(wǎng)絡(luò)的輸入樣本，并對(duì)網(wǎng)絡(luò)進(jìn)行學(xué)習(xí)訓(xùn)練，完成對(duì)刀具磨損狀態(tài)的有效識(shí)別。   關(guān)鍵詞：刀具磨損；聲發(fā)射；小波包分析；神經(jīng)網(wǎng)絡(luò)   中圖分類號(hào)：TP206＋.1 文獻(xiàn)標(biāo)識(shí)碼：A 文章編號(hào)：1672-4984(2006)02-0040-03   Acoustic emission and wavelet analysis-based estimation of tool wear   XIE Xiu-xian,FU Pan,CAO Wei-qing (Southwest Jiaotong University, Chengdu 610031，China)   Abstract：Accompanied with the development of modern machining industry, tool wear monitoring becomes more and more important. Acoustic Emission (AE) is a useful and effective technique in tool wear monitoring. This paper uses Daubechies Wavelet to analyze AE signal and select features of the tools. The selected features are then considered as inputs to BP neural network to complete recognition of the status of the cutting tool.   Key words：Tool wear; Acoustic emission; Wavelet analysis; Neural network