【摘要】：為提高發(fā)動(dòng)機(jī)轉(zhuǎn)動(dòng)部件性能衰退故障診斷精度,針對傳統(tǒng)的淺層網(wǎng)絡(luò)和支持向量機(jī)(SVM)方法在診斷時(shí)存在泛化能力欠缺、易產(chǎn)生局部最優(yōu)解等問題,引入近年來在模式識(shí)別領(lǐng)域取得巨大突破,模擬人腦多層結(jié)構(gòu)的深度置信網(wǎng)絡(luò)(DBN)進(jìn)行發(fā)動(dòng)機(jī)部件性能衰退故障的診斷。為改進(jìn)深度置信網(wǎng)絡(luò)性能,提出一種在無監(jiān)督和有監(jiān)督訓(xùn)練階段都可自適應(yīng)調(diào)整權(quán)值的改進(jìn)算法(ad_DBN)。以渦扇發(fā)動(dòng)機(jī)為對象,將兩種DBN算法與BP,RBF和SVM方法從診斷精度、計(jì)算時(shí)間、抗噪能力三方面進(jìn)行綜合比較分析。結(jié)果表明DBN算法診斷精度明顯優(yōu)于反向傳播(BP)神經(jīng)網(wǎng)絡(luò),徑向基(RBF)神經(jīng)網(wǎng)絡(luò)和支持向量機(jī)(SVM)方法,得益于權(quán)值的自適應(yīng)調(diào)整,ad_DBN診斷的平均精度高達(dá)97.84%,其抗噪聲能力也明顯優(yōu)于其他算法,能夠提高故障診斷的有效性和可靠性。
[Abstract]:In order to improve the fault diagnosis accuracy of engine rotating component performance decline, the traditional shallow layer network and support vector machine (SVM) method have some problems, such as lack of generalization ability, easy to produce local optimal solution and so on. In recent years, a great breakthrough has been made in the field of pattern recognition, which simulates the deep confidence network (DBN) of the multi-layer structure of the human brain to diagnose the performance degradation of the engine components. In order to improve the performance of deep confidence networks, an improved algorithm (ad_DBN), which adaptively adjusts weights in both unsupervised and supervised training stages, is proposed. Taking turbofan engine as an object, two kinds of DBN algorithms and BP,RBF and SVM methods are compared and analyzed comprehensively from three aspects: diagnostic accuracy, calculation time and anti-noise ability. The results show that the diagnostic accuracy of DBN algorithm is obviously better than that of backpropagation (BP) neural network, radial basis (RBF) neural network and support vector machine (SVM) method, which benefit from the adaptive adjustment of weights. The average accuracy of ad_DBN diagnosis is as high as 97.84%, and its anti-noise ability is obviously superior to other algorithms. It can improve the validity and reliability of fault diagnosis.
【作者單位】： 海軍航空工程學(xué)院飛行器工程系;海軍航空工程學(xué)院研究生管理大隊(duì);
【基金】：國家自然科學(xué)基金(51505492) “泰山學(xué)者”建設(shè)工程專項(xiàng)經(jīng)費(fèi)資助
【分類號(hào)】：V263.6

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 張立眾;孫志勇;;“逐項(xiàng)排除法”在故障診斷中的應(yīng)用[J];礦山機(jī)械;1990年09期

2 賈民平;機(jī)械故障診斷學(xué)的理論及其應(yīng)用 第一講 故障診斷的意義及研究發(fā)展概況[J];江蘇機(jī)械制造與自動(dòng)化;1999年01期

3 楊曉磊;;淺談我國鐵路機(jī)車故障診斷[J];科技風(fēng);2014年06期

4 田少民;工程機(jī)械的狀態(tài)監(jiān)測與故障診斷技術(shù)[J];工程機(jī)械;2001年01期

5 王敏,王萬俊,熊春山,黃心漢;基于多傳感器數(shù)據(jù)融合的故障診斷技術(shù)[J];華中科技大學(xué)學(xué)報(bào);2001年02期

6 潘松海;介紹一種故障診斷方法[J];組合機(jī)床與自動(dòng)化加工技術(shù);2001年06期

7 李德躍;發(fā)動(dòng)機(jī)突然熄火的故障診斷[J];城市車輛;2001年02期

8 王小虎;機(jī)械式風(fēng)速表檢定中的故障診斷及維修方法[J];中國計(jì)量;2002年03期

9 王清照,肖衛(wèi)杰,王加璇;運(yùn)用熱經(jīng)濟(jì)學(xué)結(jié)構(gòu)理論進(jìn)行故障診斷的探討[J];中國電機(jī)工程學(xué)報(bào);2003年09期

10 陳東林;煙草設(shè)備的故障診斷技術(shù)應(yīng)用與展望[J];中國設(shè)備工程;2003年06期

相關(guān)會(huì)議論文 前10條

1 楊其校;劉昭度;齊志權(quán);馬岳峰;;汽車ABS電機(jī)故障診斷[A];第三屆全國信息獲取與處理學(xué)術(shù)會(huì)議論文集[C];2005年

2 黎清海;高慶;;基于系統(tǒng)分層的故障診斷方法[A];第三屆全國信息獲取與處理學(xué)術(shù)會(huì)議論文集[C];2005年

3 聞競競;黃道;;故障診斷方法綜述[A];計(jì)算機(jī)技術(shù)與應(yīng)用進(jìn)展·2007——全國第18屆計(jì)算機(jī)技術(shù)與應(yīng)用（CACIS）學(xué)術(shù)會(huì)議論文集[C];2007年

4 李t，

本文編號(hào)：2447775