僅為學(xué)習(xí)羡玛，翻譯進(jìn)行中...?
【題目】Multiple Wavelet Regularized Deep Residual Networks for Fault Diagnosis
【翻譯】基于多組小波正則化深度殘差網(wǎng)絡(luò)的故障診斷
Abstract (摘要)
Abstract: As an emerging deep learning method, deep residual networks are gradually becoming popular in the research field of machine fault diagnosis. A significant task in deep residual network-based fault diagnosis is to prevent overfitting, which is often a major reason for low diagnostic accuracy when there is insufficient training data. This paper develops a multiple wavelet regularized deep residual network (MWR-DRN) model that uses one wavelet basis function (WBF) as the primary WBF and other WBFs as the auxiliary WBFs. “Regularized” means that a constraint or restriction is applied to yield a high performance on the testing data. To be specific, the developed MWR-DRN model is trained not only by the 2D matrices from the primary WBF, but also by the 2D matrices from the auxiliary WBFs using a stochastic selection strategy. Experimental results validate the effectiveness of the developed MWR-DRN in improving diagnostic accuracy.
【翻譯】作為一種新興的深度學(xué)習(xí)方法别智，深度殘差網(wǎng)絡(luò)在故障診斷領(lǐng)域逐漸流行起來。在基于深度殘差網(wǎng)絡(luò)的故障診斷中稼稿，一個(gè)重要的任務(wù)是避免過擬合薄榛。其中，過擬合是樣本量不足時(shí)故障診斷準(zhǔn)確率低的一個(gè)主要原因让歼。本文提出了一種多組小波正則化的深度殘差網(wǎng)絡(luò)（Multiple Wavelet Regularized Deep Residual Network敞恋，MWR-DRN），使用一個(gè)小波基函數(shù)（wavelet basis function谋右，WBF）作為主WBF硬猫，將其他WBF作為輔助WBF。具體而言改执，所提出的MWR-DRN模型不僅被主WBF的二維矩陣所訓(xùn)練啸蜜，而且被隨機(jī)選擇策略下的輔助WBF的二維矩陣所訓(xùn)練。實(shí)驗(yàn)結(jié)果驗(yàn)證了所提出MWR-DRN在提高診斷準(zhǔn)確率時(shí)的有效性辈挂。
【關(guān)鍵詞】Deep learning, deep residual learning, fault diagnosis, multiple wavelet regularization, wavelet packet transform.
【翻譯】深度學(xué)習(xí)衬横，深度殘差學(xué)習(xí)，故障診斷终蒂，多組小波閾值化蜂林，小波包變換
Reference
M. Zhao, B. Tang, L. Deng, and M. Pecht, "Multiple Wavelet Regularized Deep Residual Networks for Fault Diagnosis," Measurement, 2019.
https://www.sciencedirect.com/science/article/pii/S0263224119311959
