融合型深度學(xué)習(xí)亞健康識別算法的研究
發(fā)布時(shí)間：2019-04-08 08:44
【摘要】：在現(xiàn)代的工業(yè)生產(chǎn)過程中,逐步向生產(chǎn)設(shè)備大型化、智能化、復(fù)雜化、自動化發(fā)展,如果其中某零部件發(fā)生故障,會對整個生產(chǎn)過程產(chǎn)生一定的影響。但是若能對當(dāng)前系統(tǒng)狀態(tài)進(jìn)行準(zhǔn)確識別,并對相關(guān)故障環(huán)節(jié)的設(shè)備進(jìn)行及時(shí)的更換,就能有效的避免故障的發(fā)生。對于不可預(yù)測的外力作用所導(dǎo)致的突發(fā)性系統(tǒng)故障來說一般是沒有任何征兆以及不可控的。但是工業(yè)過程控制系統(tǒng)發(fā)生的故障大多是由于設(shè)備磨損或者元件老化導(dǎo)致的延時(shí)性故障,表征為設(shè)備工作時(shí)表現(xiàn)不明顯,因此延時(shí)性故障研究和設(shè)備可靠性研究成為工業(yè)設(shè)備故障診斷領(lǐng)域的重點(diǎn),并引起很多的專家學(xué)者的關(guān)注。本文主要對深度自動編碼器及其改進(jìn)進(jìn)行研究,并以滾動軸承的“亞健康”狀態(tài)識別作為應(yīng)用場景。在閱讀大量的深度學(xué)習(xí)和故障診斷方法后,發(fā)現(xiàn)一個設(shè)計(jì)良好的學(xué)習(xí)率策略可以顯著提高深度學(xué)習(xí)模型的收斂速度,于是本文提出一種自適應(yīng)性學(xué)習(xí)率來提高深度網(wǎng)絡(luò)模型的收斂速度。同時(shí)本文結(jié)合稀疏自動編碼器以及邊緣降噪編碼器的優(yōu)點(diǎn),對原深度模型的代價(jià)函數(shù)進(jìn)行改進(jìn),增加網(wǎng)絡(luò)模型的泛化能力以及魯棒性。本文采用層疊編碼機(jī)作為深度學(xué)習(xí)的網(wǎng)絡(luò)結(jié)構(gòu),通過這種結(jié)構(gòu)可以對機(jī)械振動信號中的噪聲進(jìn)行過濾,有助于有利特征的提取。實(shí)驗(yàn)結(jié)果表明在基本保證準(zhǔn)確率的情況下加快了模型的收斂速度。本文使用相關(guān)向量機(jī)取代傳統(tǒng)深度學(xué)習(xí)中的SoftMax層對提取的深度特征進(jìn)行識別分類,同時(shí),相關(guān)向量機(jī)的核函數(shù)選取以及核參數(shù)的選取對于最終的分類結(jié)果尤為重要,本文采用混合模式的核函數(shù),并且根據(jù)Fisher準(zhǔn)則以及最大熵準(zhǔn)則提出一種最優(yōu)映射的相關(guān)向量機(jī)核參數(shù)選取方法。實(shí)驗(yàn)結(jié)果表明這種方式下選取的核參數(shù)有助于提高模型的識別精度。為了進(jìn)一步提高識別的精確性,本文將由相關(guān)向量機(jī)分類器所得到的輸出歸一化處理后作為D-S證據(jù)理論的第一證據(jù)空間,將由SoftMax分類器得到的結(jié)果歸一化后作為第二個證據(jù)空間,然后根據(jù)D-S證據(jù)理論的融合規(guī)則將兩個證據(jù)空間進(jìn)行融合得到最終的識別結(jié)果。實(shí)驗(yàn)結(jié)果表明該方法在提高滾動軸承“亞健康”狀態(tài)識別精度上是有效的。
[Abstract]:In the process of modern industrial production, the production equipment is gradually developed to large-scale, intelligent, complex and automatic. If one of the parts is broken down, it will have a certain impact on the whole production process. However, if the current system status can be accurately identified, and the equipment of the relevant fault links can be replaced in a timely manner, the failure can be effectively avoided. For unexpected system failures caused by unpredictable external forces, there are generally no signs and uncontrollable. However, most of the industrial process control system failures are caused by equipment wear or aging, which is characterized as the performance of the equipment is not obvious. Therefore, the research on delay fault and equipment reliability has become the focal point in the field of industrial equipment fault diagnosis, and has attracted the attention of many experts and scholars. This paper mainly studies the depth automatic encoder and its improvement, and takes the "sub-health" state recognition of rolling bearings as the application scenario. After reading a large number of in-depth learning and fault diagnosis methods, it is found that a well-designed learning rate strategy can significantly improve the convergence rate of the deep learning model. Therefore, this paper proposes a self-adaptive learning rate to improve the convergence rate of the deep network model. At the same time, combining the advantages of sparse automatic encoder and edge de-noising encoder, the cost function of the original depth model is improved to increase the generalization ability and robustness of the network model. In this paper, the cascade coding machine is used as the network structure of in-depth learning. Through this structure, the noise in mechanical vibration signal can be filtered, which is helpful to the extraction of favorable features. The experimental results show that the convergence speed of the model is speeded up under the condition that the accuracy of the model is basically guaranteed. In this paper, we use the correlation vector machine to replace the SoftMax layer in the traditional depth learning to identify and classify the extracted depth features. At the same time, the kernel function selection and the kernel parameter selection of the correlation vector machine are very important for the final classification results. In this paper, the kernel function of mixed mode is used, and according to the Fisher criterion and the maximum entropy criterion, a method for selecting the kernel parameters of the correlation vector machine is proposed. The experimental results show that the kernel parameters selected in this way are helpful to improve the recognition accuracy of the model. In order to further improve the accuracy of recognition, the output normalization obtained from the correlation vector machine classifier is regarded as the first evidence space of DES evidence theory. The results obtained by SoftMax classifier are normalized as the second evidence space, and then the two evidence spaces are fused according to the fusion rule of DES evidence theory to get the final recognition result. The experimental results show that this method is effective in improving the accuracy of "sub-health" state identification of rolling bearings.
【學(xué)位授予單位】：遼寧大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP18
本文編號：2454445
[Abstract]:In the process of modern industrial production, the production equipment is gradually developed to large-scale, intelligent, complex and automatic. If one of the parts is broken down, it will have a certain impact on the whole production process. However, if the current system status can be accurately identified, and the equipment of the relevant fault links can be replaced in a timely manner, the failure can be effectively avoided. For unexpected system failures caused by unpredictable external forces, there are generally no signs and uncontrollable. However, most of the industrial process control system failures are caused by equipment wear or aging, which is characterized as the performance of the equipment is not obvious. Therefore, the research on delay fault and equipment reliability has become the focal point in the field of industrial equipment fault diagnosis, and has attracted the attention of many experts and scholars. This paper mainly studies the depth automatic encoder and its improvement, and takes the "sub-health" state recognition of rolling bearings as the application scenario. After reading a large number of in-depth learning and fault diagnosis methods, it is found that a well-designed learning rate strategy can significantly improve the convergence rate of the deep learning model. Therefore, this paper proposes a self-adaptive learning rate to improve the convergence rate of the deep network model. At the same time, combining the advantages of sparse automatic encoder and edge de-noising encoder, the cost function of the original depth model is improved to increase the generalization ability and robustness of the network model. In this paper, the cascade coding machine is used as the network structure of in-depth learning. Through this structure, the noise in mechanical vibration signal can be filtered, which is helpful to the extraction of favorable features. The experimental results show that the convergence speed of the model is speeded up under the condition that the accuracy of the model is basically guaranteed. In this paper, we use the correlation vector machine to replace the SoftMax layer in the traditional depth learning to identify and classify the extracted depth features. At the same time, the kernel function selection and the kernel parameter selection of the correlation vector machine are very important for the final classification results. In this paper, the kernel function of mixed mode is used, and according to the Fisher criterion and the maximum entropy criterion, a method for selecting the kernel parameters of the correlation vector machine is proposed. The experimental results show that the kernel parameters selected in this way are helpful to improve the recognition accuracy of the model. In order to further improve the accuracy of recognition, the output normalization obtained from the correlation vector machine classifier is regarded as the first evidence space of DES evidence theory. The results obtained by SoftMax classifier are normalized as the second evidence space, and then the two evidence spaces are fused according to the fusion rule of DES evidence theory to get the final recognition result. The experimental results show that this method is effective in improving the accuracy of "sub-health" state identification of rolling bearings.
【學(xué)位授予單位】：遼寧大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP18
【參考文獻(xiàn)】
相關(guān)期刊論文 前5條
1 宋暉;薛云;張良均;;基于SVM分類問題的核函數(shù)選擇仿真研究[J];計(jì)算機(jī)與現(xiàn)代化;2011年08期
2 曹正洪;沈繼紅;;基于模糊集理論的傳感器健康度評價(jià)方法[J];電機(jī)與控制學(xué)報(bào);2010年05期
3 王維民;高金吉;李燕;李雙喜;;離心壓縮機(jī)軸位移故障自愈調(diào)控系統(tǒng)試驗(yàn)研究[J];機(jī)械工程學(xué)報(bào);2010年05期
4 戚曉利;潘紫微;;自適應(yīng)遺傳算法在滾動軸承故障診斷中的應(yīng)用[J];機(jī)械傳動;2006年05期
5 高金吉;裝備系統(tǒng)故障自愈原理研究[J];中國工程科學(xué);2005年05期
相關(guān)博士學(xué)位論文 前4條
1 李海生;基于證據(jù)理論的分類方法研究[D];華南理工大學(xué);2013年
2 陳渤;基于核方法的雷達(dá)高分辨距離像目標(biāo)識別技術(shù)研究[D];西安電子科技大學(xué);2008年
3 于雪蓮;基于核方法和流形學(xué)習(xí)的雷達(dá)目標(biāo)距離像識別研究[D];電子科技大學(xué);2008年
4 任雙橋;支撐矢量機(jī)理論與應(yīng)用研究[D];國防科學(xué)技術(shù)大學(xué);2006年
相關(guān)碩士學(xué)位論文 前3條
1 楊永波;基于機(jī)器學(xué)習(xí)的故障狀態(tài)評價(jià)算法研究[D];遼寧大學(xué);2013年
2 黃嘯;支持向量機(jī)核函數(shù)的研究[D];蘇州大學(xué);2008年
3 邊寶峰;基于神經(jīng)網(wǎng)絡(luò)和D-S證據(jù)理論的信息融合方法的研究[D];華北電力大學(xué)（河北）;2005年
，本文編號：2454445
本文鏈接：http://sikaile.net/kejilunwen/zidonghuakongzhilunwen/2454445.html
最近更新
教材專著
