本文關(guān)鍵詞：基于PSO聚類和特征貢獻(xiàn)度的油液監(jiān)測(cè)信息特征選擇方法　出處：《潤(rùn)滑與密封》2016年01期 　論文類型：期刊論文

  更多相關(guān)文章： 機(jī)械裝備 油液監(jiān)測(cè) 特征選擇 粒子群聚類 特征貢獻(xiàn)度

【摘要】：特征選擇是實(shí)現(xiàn)油液監(jiān)測(cè)多技術(shù)手段綜合應(yīng)用的關(guān)鍵問題之一。針對(duì)油液監(jiān)測(cè)信息特點(diǎn),提出一種油液監(jiān)測(cè)信息特征選擇方法。該方法首先采用K均值PSO聚類算法對(duì)樣本實(shí)施無(wú)監(jiān)督聚類,實(shí)現(xiàn)樣本的預(yù)先分類;然后采用定義的特征貢獻(xiàn)度,計(jì)算各特征對(duì)聚類結(jié)果的貢獻(xiàn)度,并以此作為特征選擇的依據(jù),實(shí)現(xiàn)無(wú)監(jiān)督的過(guò)濾式特征選擇。通過(guò)在某型柴油機(jī)潤(rùn)滑油原子發(fā)射光譜和紅外光譜信息中的應(yīng)用表明,該算法能夠很好的實(shí)現(xiàn)油液監(jiān)測(cè)信息的特征選擇,減少特征指標(biāo)數(shù)量,而且能夠避免由于油液監(jiān)測(cè)信息依存度和相關(guān)度高的特點(diǎn)而造成特征選擇時(shí)可能會(huì)將重要信息刪除的問題。
[Abstract]:Feature selection is one of the key problems in realization of oil monitoring and comprehensive application of multiple technologies. According to the characteristics of oil monitoring information, proposes a feature selection method of oil liquid monitoring information. The method first uses K PSO means clustering algorithm to sample the implementation of unsupervised clustering, sample pre classification; then the definition of feature contribution degree the contribution of each feature, the calculation of the clustering results, and as a feature selection based on selection of filter characteristics. Through unsupervised in a certain type of diesel engine lubricating oil by atomic emission spectroscopy and infrared spectral information shows that this algorithm can achieve very good characteristics of oil monitoring information, reduce the number of features, and can avoid the oil monitoring information dependency degree and high correlation characteristics caused by feature selection may issue important information will be deleted.

【作者單位】： 海軍工程大學(xué)青島油液檢測(cè)分析中心;武漢理工大學(xué)能源與動(dòng)力工程學(xué)院;
【基金】：國(guó)家自然科學(xué)基金項(xiàng)目(51309185)
【分類號(hào)】：TK421.9
【正文快照】： j縥縥縥縥縥縥，

本文編號(hào)：1398987