程曉涵 汪愛明 蘇一新 孟國營

摘要： 提出了一種融合投影尋蹤的自回歸分析方法實(shí)現(xiàn)設(shè)備的預(yù)知維護(hù)。該預(yù)知維護(hù)方法是從設(shè)備關(guān)鍵部件處提取振動(dòng)信號(hào)，經(jīng)分析和計(jì)算得出24種特征指標(biāo)用以描述設(shè)備運(yùn)行狀態(tài)；對(duì)24種特征指標(biāo)分別提取一個(gè)時(shí)間序列并各自進(jìn)行自回歸分析，得到各自對(duì)應(yīng)的預(yù)測因子；利用投影尋蹤將前述預(yù)測因子投影到二維空間，然后分別建立預(yù)測因子投影值與相對(duì)應(yīng)的特征指標(biāo)值的擬合函數(shù)，進(jìn)而推算出24種特征指標(biāo)的未來值；再通過對(duì)24種特征指標(biāo)的未來值在最佳投影方向矩陣下進(jìn)行投影，根據(jù)投影值的分布情況判斷設(shè)備未來運(yùn)行狀態(tài)是否存在異常，從而實(shí)現(xiàn)設(shè)備的預(yù)知維護(hù)。最后利用美國西儲(chǔ)大學(xué)軸承數(shù)據(jù)中心網(wǎng)站公開發(fā)布的軸承探傷測試數(shù)據(jù)集中的內(nèi)圈故障數(shù)據(jù)和山西省某洗煤廠主井皮帶機(jī)的減速器故障數(shù)據(jù)進(jìn)行了驗(yàn)證。關(guān)鍵詞： 故障診斷； 特征指標(biāo)； 投影尋蹤； 評(píng)價(jià)體系； 預(yù)知維護(hù)方法

中圖分類號(hào)： TH165+.3； TP277文獻(xiàn)標(biāo)志碼： A文章編號(hào)： 10044523（2016）04063107

DOI：10.16385/j.cnki.issn.10044523.2016.04.010

引言

目前設(shè)備的日常運(yùn)行和維護(hù)工作基本上是依托于各種在線或者離線的監(jiān)測監(jiān)控系統(tǒng)，并主要依靠門限報(bào)警方式，通過對(duì)不同的監(jiān)測參數(shù)設(shè)置一定的極限值進(jìn)行超值報(bào)警，但當(dāng)達(dá)到報(bào)警值時(shí)設(shè)備已經(jīng)出現(xiàn)了不同程度的損壞或性能降低，導(dǎo)致設(shè)備存在很大的安全隱患，并使維護(hù)維修工作存在嚴(yán)重的被動(dòng)性，因此需要尋求方法實(shí)現(xiàn)設(shè)備運(yùn)行狀態(tài)的預(yù)知，及時(shí)采取維護(hù)措施。

設(shè)備運(yùn)行狀況是否正常可以由不同的特征指標(biāo)進(jìn)行衡量，不同的特征指標(biāo)對(duì)各類故障類型和不同的損傷程度可能表現(xiàn)出不同的靈敏度和響應(yīng)特點(diǎn)，且相關(guān)性難以界定[12]，因此文章選取了24種特征指標(biāo)對(duì)設(shè)備狀態(tài)進(jìn)行描述，提出了一種基于投影尋蹤方法的設(shè)備預(yù)知維護(hù)方法，即采用基于24種特征指標(biāo)的融合投影尋蹤方法的自回歸分析，實(shí)現(xiàn)了對(duì)設(shè)備零部件未來狀態(tài)的初步預(yù)測，并根據(jù)預(yù)測結(jié)果提出相應(yīng)的維護(hù)維修建議。

Abstract： Equipment maintenance is passive when using traditional threshold alarm， so early detection of fault and prescient maintenance are badly in need. A method of regression analysis combined with projection pursuit method for equipments is put forward. The first step of this method is analyzing the vibration signal and calculating 24 characteristic indexes to describe the equipment running status. By extracting a time series respectively of these 24 characteristic indexes and regression analysis， predictors for each characteristic index are obtained. Predictors are respectively projected to twodimensional space using projection pursuit method， and then fitting functions between projection values of predictors and characteristic index are established， hereunder future values of 24 characteristic indexes are inferred. Then projecting these 24 future values under the best projection direction matrix and observation of distribution of projection values， whether the future status of the equipment abnormal could be judged and prescient maintenance of coal mine equipment could be realized. At last， the method proposed in this paper is verified by inner ring fault test data publicly released by the Case Western Reserve University Bearing Data Center website and failure data when gear reducer served to belt convey at main shaft of coal washery of one mining group in Shanxi Province was broken.

Key words： fault diagnosis； characteristic indexes； projection pursuit method； evaluation index system； predictive maintenance method