检索结果-内蒙古大学图书馆

Enhancing unsupervised fault diagnosis method for rolling bearings using domain adaptive transfer coding techniques

MEASUREMENT SCIENCE AND TECHNOLOGY 2025年第6期36卷 066112-066112页

作者： Lu, Jiantao Xiao, Zhilin Zhang, Peng Li, Shunming Nanjing Univ Aeronaut & Astronaut Sch Energy & Power Nanjing 210016 Peoples R China AECC Guiyang Engine Design Res Inst Guiyang 550081 Peoples R China Nantong Inst Technol Sch Automot Engn Nantong 226002 Peoples R China

Intelligent fault diagnosis of rolling bearings in unsupervised conditions remains a great challenge. Transfer learning plays a crucial role in addressing this issue by leveraging knowledge gained from labeled source domain data to enhance the diagnostic performance on unlabeled target domain samples. However, traditional transfer diagnosis techniques often face significant challenges due to missing labels in target domain data under variable operating conditions and considerable discrepancies regarding the distribution of data between target and source domain. These issues can substantially degrade the performance of transfer diagnosis techniques. To address these issues, this study introduces a domain adaptive-based attention group convolution transfer network (DA-AGCTN) specifically for rolling bearing fault diagnosis using unlabeled samples in diverse operational scenarios. The proposed approach consists of two main components, namely DA label generation module (DALGM) and AGCTN. DALGM utilizes a synergistic enhanced stacked autoencoder (SAE) to enhance pseudo label generation through robust feature extraction from unlabeled target domain data. This optimization, which includes a novel domain confusion metric that combines reconstruction loss, domain distinction loss and probabilistic classification loss, is designed to extract domain-invariant features more effectively. Following feature extraction by SAE and dimensionality reduction via t-distribution stochastic neighborhood embedding (abbreviated to t-SNE), the K-means method is employed for clustering, with a subsequent approach for aligning labels to create more accurate pseudo labels. AGCTN integrates attention mechanism and group convolution to capture shared features efficiently, leveraging a pretrained source domain model to enhance target domain generalization. The effectiveness of the proposed DA-AGCTN is corroborated by rolling bearing fault simulation experiments, demonstrating superior diagnostic a

关键词： stacked autoencoder bearing fault diagnosis attention mechanism unsupervised technology transfer learning

来源：评论

学校读者我要写书评

暂无评论

High-power disk laser welding statuses monitoring based on analyses of multiple-sensor signals

引用

JOURNAL OF MANUFACTURING PROCESSES 2019年第0期41卷 221-230页

作者： Zhang, Yanxi Zhang, Nanfeng You, Deyong Gao, Xiangdong Katayama, Seiji Guangdong Univ Technol Guangdong Prov Welding Engn Technol Res Ctr Guangzhou 510006 Guangdong Peoples R China Osaka Univ Joining & Welding Res Inst 11-1 Mihogaoka Ibaraki Osaka 5670047 Japan

Welding statuses monitoring is crucial to quality control during high power disk laser welding of thick plates. A multiple-sensor-system is designed to capture the features of the keyhole, plume, spatter, optical and spectrum information of the welding area. These signals comprehensively depict the welding statuses, and are used to online monitor the welding statuses. The correlations between the captured multiple signals are analyzed by the correlation analysis method and Linear discriminative analysis (LDA) analysis and stacked Auto-Encoder (SAE) are implemented, and the dimension-reduced non-linear transformation of the original features acquired by SAE shows better discriminative and representative capacity than the linear combination of the original features acquired by the LDA. This research not only investigates the correlations between the signals of the keyhole, plume, photodiode and spectrum information during high power disc laser welding but also provides a novel method to conduct online welding statuses monitoring. Three different welding experiments under different parameters were conducted, and these experiments include blowout, humping and undercut defects, respectively.

关键词： Welding statuses monitoring Correlation analysis Linear discriminative analysis stacked autoencoder Multiple-sensor

来源：评论

学校读者我要写书评

暂无评论

Physiological-signal-based mental workload estimation via transfer dynamical autoencoders in a deep learning framework

引用

NEUROCOMPUTING 2019年 347卷 212-229页

作者： Yin, Zhong Zhao, Mengyuan Zhang, Wei Wang, Yongxiong Wang, Yagang Zhang, Jianhua Univ Shanghai Sci & Technol Engn Res Ctr Opt Instrument & Syst Shanghai Key Lab Modern Opt Syst Minist Educ Shanghai 200093 Peoples R China Univ Shanghai Sci & Technol Sch Marxism Shanghai 200093 Peoples R China Oslo Metropolitan Univ Dept Comp Sci OsloMet Artificial Intelligence Lab N-0130 Oslo Norway

Evaluating operator mental workload (MW) in human-machine systems via neurophysiological signals is crucial for preventing unpredicted operator performance degradation. However, the feature of physiological signals is associated with the historical values at the previous time steps and its statistical properties vary across individuals and types of mental tasks. In this study, we propose a new transfer dynamical autoencoder (TDAE) to capture the dynamical properties of electroencephalograph (EEG) features and the individual differences. The TDAE consists of three consecutively-connected modules, which are termed as feature filter, abstraction filter, and transferred MW classifier. The feature and abstraction filters introduce dynamical deep network to abstract the EEG features across adjacent time steps to salient MW indicators. Transferred MW classifier exploits large volume EEG data from an source-domain EEG database recorded under emotional stimuli to improve the model training stability. We tested our algorithms on two target EEG databases. The classification performance shows TDAE significantly outperforms existing shallow and deep MW classification models. We also investigated how to select TDAE hyper-parameters and found its superiority in accuracy can be achieved with proper filter orders. (C) 2019 Elsevier B.V. All rights reserved.

关键词： Mental workload Electroencephalograph Deep learning stacked autoencoder Operator functional state

来源：评论

学校读者我要写书评

暂无评论

A Learning-Based Multimodel Integrated Framework for Dynamic Traffic Flow Forecasting

引用

NEURAL PROCESSING LETTERS 2019年第1期49卷 407-430页

作者： Zhou, Teng Han, Guoqiang Xu, Xuemiao Han, Chu Huang, Yuchang Qin, Jing Shantou Univ Dept Comp Sci Coll Engn Shantou Peoples R China South China Univ Technol Sch Comp Sci & Engn Guangzhou Guangdong Peoples R China Chinese Univ Hong Kong Dept Comp Sci & Engn Sha Tin Hong Kong Peoples R China South China Agr Univ Coll Math & Informat Guangzhou Guangdong Peoples R China Hong Kong Polytech Univ Ctr Smart Hlth Sch Nursing Kowloon Hong Kong Peoples R China

Accurate and timely traffic flow forecasting is essential for many intelligent transportation systems. However, it is quite challenging to develop an efficient and robust forecasting model due to the inherent randomness and large variations of traffic flow. Over the past two decades, a variety of traffic flow forecasting models have been proposed. While each model has its merits and can achieve satisfactory forecasting results under certain traffic conditions, it is difficult for a single model to deal with various conditions well. In this paper, we proposed a novel deep learning-based multimodel integration framework in order to overcome the limitations of previous methods in dealing with large variations and uncertainties of traffic flow and hence improve the forecasting accuracy. Our framework can dynamically choose an optimal model or an optimal subset of models from a set of candidate models to forecast the future traffic flow conditions according to current input data. We employ stacked autoencoder (SAE), a simple yet efficient deep learning architecture, to extract the implicit relationships hidden in the traffic flow data and employed labeled data to fine tune the parameters of the architecture. Compared with the hand-crafted features and explicable dependence relations leveraged in previous models, the features learning from SAE are more representative and hence have more powerful forecasting capability. In addition, we propose a model-driven scheme to automatically label the training data and develop three strategies to integrate multiple models. Extensive experiments performed on three typical traffic flow datasets demonstrate the proposed framework outperforms state-of-the-art models and achieves much more accurate forecasting results under large and sudden variations.

关键词： Traffic flow forecasting stacked autoencoder Multimodel integration Variation and uncertainty Deep learning

来源：评论

学校读者我要写书评

暂无评论

A Novel Algorithm for High-Resolution Magnetic Induction Tomography Based on stacked Auto-Encoder for Biological Tissue Imaging

引用

IEEE ACCESS 2019年 7卷 185597-185606页

作者： Chen, Ruijuan Huang, Juan Wang, Huiquan Li, Bingnan Zhao, Zhe Wang, Jinhai Wang, Yao Tianjin Polytech Univ Sch Life Sci Tianjin 300387 Peoples R China Tianjin Key Lab Optoelect Detect Technol & Syst Tianjin 300387 Peoples R China

Magnetic induction tomography (MIT) is a non-invasive and non-contact imaging method that uses an excitation coil to generate time-varying magnetic fields in space and reconstruct the internal conductivity distribution based on the phase difference. In this study, a new MIT reconstruction algorithm was proposed for non-contact measurement and monitoring of the location of the anomaly in the biomedical object of interest. To reconstruct the distribution of electrical characteristics inside the biological tissue, this technique uses a stacked auto-encoder (SAE) neural network composed of a multi-layer automatic encoder. The location and reconstruction accuracy of the anomaly based on SAE and back-projection were compared, and a hemorrhagic stroke was simulated to verify the practicability of the proposed algorithm. The results showed that the relative error of reconstruction based on the SAE network algorithm reached 0.29%, which improved anomaly reconstruction accuracy and reduced the prediction time to 0.02 s. At the same time, the network was used for the reconstruction of hemorrhagic stroke in different locations, amounts, and shapes. Accordingly, the SAE neural network reconstruction algorithm proposed in this study, which can autonomously learn the non-linear relationship between input and output, can solve the defects of the traditional reconstruction algorithm, such as serious artifacts and complex calculations.

关键词： Electrical conductivity eddy current magnetic induction tomography stacked autoencoder reconstruction algorithm

来源：评论

学校读者我要写书评

暂无评论

Transfer Learning for Molecular Cancer Classification Using Deep Neural Networks

引用

IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS 2019年第6期16卷 2089-2100页

作者： Sevakula, Rahul K. Singh, Vikas Verma, Nishchal K. Kumar, Chandan Cui, Yan Indian Inst Technol Kanpur Dept Elect Engn Kanpur 208016 Uttar Pradesh India Indian Inst Technol Guwahati Gauhati 781039 Assam India Univ Tennessee Ctr Hlth Sci Dept Genet Genom & Informat Memphis TN 38163 USA

The emergence of deep learning has impacted numerous machine learning based applications and research. The reason for its success lies in two main advantages: 1) it provides the ability to learn very complex non-linear relationships between features and 2) it allows one to leverage information from unlabeled data that does not belong to the problem being handled. This paper presents a transfer learning procedure for cancer classification, which uses feature selection and normalization techniques in conjunction with s sparse auto-encoders on gene expression data. While classifying any two tumor types, data of other tumor types were used in unsupervised manner to improve the feature representation. The performance of our algorithm was tested on 36 two-class benchmark datasets from the GEMLeR repository. On performing statistical tests, it is clearly ascertained that our algorithm statistically outperforms several generally used cancer classification approaches. The deep learning based molecular disease classification can be used to guide decisions made on the diagnosis and treatment of diseases, and therefore may have important applications in precision medicine.

关键词： Transfer learning cancer classification deep neural network stacked autoencoder feature selection

来源：评论

学校读者我要写书评

暂无评论

A spatial temporal neighborhood preserving method for feature learning with an industrial application

A spatial temporal neighborhood preserving method for featur...

引用

第32届中国过程控制会议(CPCC2021)

作者： Chenliang Liu Yalin Wang Kai Wang Xiaofeng Yuan School of Automation Central South University

Modern industrial process data often exhibit nonlinear and dynamic *** deep learning methods,such as stacked autoencoder(SAE),have excellent nonlinear feature learning capabilities,but they ignore the dynamic correlation between process *** learning based on manifold learning using neighborhood structure preserving has been widely used in industrial dynamic process ***,most of them extract linear features and the complex nonlinearities in process data are largely ***,a spatial temporal neighborhood preserving stack autoencoder(STNP-SAE) is proposed to learn static neighborhood features and dynamic neighborhood features of process data simultaneously in this ***,STNP-SAE is utilized to construct a soft sensor framework for quality *** effectiveness and prediction performance of the proposed method are validated on a practical hydrocracking process.

关键词： Spatial temporal neighborhood preserving feature learning stacked autoencoder quality prediction

来源：评论

学校读者我要写书评

暂无评论

The Forecasting of PM2.5 Using a Hybrid Model Based on Wavelet Transform and an Improved Deep Learning Algorithm

引用

IEEE ACCESS 2019年 7卷 142814-142825页

作者： Qiao, Weibiao Tian, Wencai Tian, Yu Yang, Quan Wang, Yining Zhang, Jianzhuang North China Univ Water Resources & Elect Power Sch Environm & Municipal Engn Zhengzhou 450046 Henan Peoples R China Sinopec Zhongyuan Petr Engn Co Ltd Zhengzhou 450046 Henan Peoples R China

In recent years, the haze has caused serious troubles to peoples lives, with the continuous increase of PM2.5 emissions. The accurate prediction of PM2.5 is very crucial for policy makers to make predictive measures. Due to the nonlinearity of the PM2.5 time series, it is difficult to predict accurately. Despite some studies about PM2.5 being proposed, the problem of the LSTM (long short-term memory) gradient disappearance and random selection of wavelet orders and layers isnt still solved. In this study, a novel model based on WT (wavelet transform)-SAE (stacked autoencoder)-LSTM is proposed. Firstly, six study sites from China are taken as examples and WT is used to decompose PM2.5 time series into several low-and high- frequency components based on different samples. Secondly, the decomposed components are predicted based on SAE-LSTM. Finally, the predicted results are reconstructed in view of all low-and high-frequency components and the predicted results are obtained. The results imply that: (1) the forecasting performance of SAE-LSTM is better than that of other models (e.g., BP (back propagation)) used for comparison;(2) for six different PM 2.5 samples, four orders five layers, five orders six layers, five orders seven layers, three orders six layers, five orders seven layers, and five orders six layers are the most appropriate. The conclusion that such a novel model may help to enhance the accuracy of PM 2.5 prediction can be drawn.

关键词： Predictive models Forecasting Wavelet transforms Logic gates Neural networks Atmospheric modeling PM 2 5 time series wavelet transform stacked autoencoder long short-term memory prediction

来源：评论

学校读者我要写书评

暂无评论

An Efficient Framework for Animal Breeds Classification Using Semi-Supervised Learning and Multi-Part Convolutional Neural Network (MP-CNN)

引用

IEEE ACCESS 2019年 7卷 151783-151802页

作者： Meena, S. Divya Agilandeeswari, L. Vellore Inst Technol Sch Informat Technol & Engn Vellore 632014 Tamil Nadu India

The automatic classification of animal images is an onerous task due to the challenging image conditions, especially when it comes to animal breeds. In this paper, we built a semi-supervised learning based Multi-part Convolutional Neural Network (MP-CNN) that classifies 35,992 animal images from ImageNet into 27 different classes of animals. The proposed model classifies the animals on both generic and fine-grained level. The animal breeds are accurately classified using Multi-part Convolutional Neural Network with a hybrid feature extraction framework of Fisher Vector based stacked autoencoder. Furthermore, with Semi-supervised learning based pseudo-labels, the model classifies new classes of unlabeled images too. Modified Hellinger Kernel classifier has been used to re-train the misclassified classes of animals and thereby improve the performance obtained from MP-CNN. The model has experimented with varied tasks to analyze its performance in each of the cases. The experimental results have proved that the coalesced approach of MP-CNN with pseudo-labels can accurately classify animal breeds and we have achieved an accuracy of 99.95 from the proposed model.

关键词： Feature extraction Convolutional neural networks Dogs Semisupervised learning Image classification Fisher vector inception-V3 modified hellinger Kernel classifier multi-part based convolutional neural network pseudo-labels semi-supervised learning stacked autoencoder

来源：评论

学校读者我要写书评

暂无评论

Intelligent fall detection method based on accelerometer data from a wrist-worn smart watch

引用

MEASUREMENT 2019年 140卷 215-226页

作者： Chen, Lin Li, Rong Zhang, Hang Tian, Lili Chen, Ning Hunan Univ State Key Lab Adv Design & Mfg Vehicle Body Changsha 410082 Hunan Peoples R China Hunan Univ Coll Mech & Vehicle Engn Changsha 410082 Hunan Peoples R China

Accurate and reliable automatic fall detection based on wearable devices enables elderly people to receive instant treatment and can alleviate the severe consequences of falls. Falls are abnormal activities that occur rarely compared with normal daily activities;therefore, fall detection can be considered a one-class classification problem. However, it is difficult and dangerous to collect sufficient fall data in practice, making it difficult to use supervised learning methods to detect falls automatically. Among wearable devices, wrist-worn devices for fall detection are more likely to be accepted because of comfort;however, high accuracy cannot typically be realized due to sensitivity to interference from the diverse activities of the hand and wrist. Combining ensemble stacked autoencoders (ESAEs) with one-class classification based on the convex hull (OCCCH), this paper proposes a novel intelligent fall detection method, namely, ESAEs-OCCCH, which is based on accelerometer data from a wrist-worn smart watch. In the proposed method, ESAEs are first adopted for unsupervised feature extraction to overcome the disadvantages of artificial feature extraction, namely, the requirements in terms of experience and time. Then, OCCCH is used for pattern recognition. Finally, the majority voting strategy and weight adaptive adjustment strategy are combined to improve the performance and stability of fall detection. According to the behavioral characteristics of the elderly, the uncertainty of activities of daily living (ADLs) and fall activities (FAs), and the influence of the intense activities of the hand on the accelerometer signal, thirteen FAs and sixteen ADLs, including intense hand and wrist activities, are simulated by young volunteers of various genders, ages, heights and weights in two experiments. The experimental results demonstrate the performance and stability of the proposed method. (C) 2019 Elsevier Ltd. All rights reserved.

关键词： Fall detection Wearable device stacked autoencoder One-class classification based on the convex hull Ensemble strategy

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：