检索结果-内蒙古大学图书馆

Gas path fault detection and isolation for aero-engine based on LSTM-DAE approach under multiple-model architecture

MEASUREMENT 2022年 202卷

作者： Wang, Kun Guo, Yingqing Zhao, Wanli Zhou, Qifan Guo, Pengfei Northwestern Polytech Univ Sch Energy & Power Xian 710129 Peoples R China

Gas path fault diagnosis plays a critical role in the security guarantee and maintenance of aero-engines. In this paper, an approach based on a fusion neural network under multiple-model architecture for gas path fault detection and isolation is proposed. We develop a multi-channel long short-term memory network based on a sliding window to explore temporal and spatial relationships of data and capture the residuals of sensor mea-surements between predicted and observed values. Additionally, denoising autoencoders under a multiple-model architecture are introduced so as to perform fault detection and isolation based on the comparison of recon-structed prediction errors and isolation thresholds. Several simulation results verify that the diagnostic model has excellent robustness and diagnostic ability. The proposed method is compared with other common methods, and the advantages and functions of this method are presented.

关键词： Aero-engine Gas path fault diagnosis Long short-term memory network denoising autoencoder Multiple -model

来源：评论

学校读者我要写书评

暂无评论

Derivation of Kinetic Parameters and Lignocellulosic Composition From Thermogram of Biomass Pyrolysis Using Convolutional Neural Network

引用

INTERNATIONAL JOURNAL OF ENERGY RESEARCH 2024年第1期2024卷

作者： Kim, Heeyoon Jo, Hyunbin Ryu, Changkook Sungkyunkwan Univ Sch Mech Engn Suwon 16419 South Korea Korea Inst Energy Res KIER Clean Fuel Res Lab Daejeon 34129 South Korea

A novel method employing a 1-dimensional convolutional neural network (1D-CNN) has been developed to deduce kinetic parameters for the three-parallel-reaction model (TPRM) and the lignocellulosic composition from the thermogram of biomass pyrolysis. This model was trained on differential thermogram (DTG) datasets created at various heating rates with rate constants randomly selected from expansive ranges. Furthermore, to enhance prediction accuracy, a denoising autoencoder (DAE) was crafted to eliminate noise from experimental data effectively. The 1D-CNN regression model forecasted kinetic parameters with mean errors of 1.52% for trained heating rates and 1.39%-3.19% for other heating rates. When tested on four biomass samples, the model precisely mimicked the DTG curves with R2 values ranging from 0.9956 to 0.9994. Relative to conventional numerical methods, this model delivers comparable prediction accuracy but through a significantly streamlined and expedited process. Enhancements are needed to broaden the model's applicability across various kinetic models and materials.

关键词： biomass convolutional neural network denoising autoencoder kinetics pyrolysis three-parallel-reaction model

来源：评论

学校读者我要写书评

暂无评论

A Building Energy Consumption Prediction Method Based on Integration of a Deep Neural Network and Transfer Reinforcement Learning

引用

INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE 2020年第10期34卷

作者： Fu, Qiming Liu, QingSong Gao, Zhen Wu, Hongjie Fu, Baochuan Chen, Jianping Suzhou Univ Sci & Technol Inst Elect & Informat Engn Suzhou 215009 Jiangsu Peoples R China Suzhou Univ Sci & Technol Jiangsu Key Lab Intelligent Bldg Energy Efficienc Suzhou 215009 Jiangsu Peoples R China McMaster Univ Fac Engn Hamilton ON L8S 0A3 Canada

With respect to the problem of the low accuracy of traditional building energy prediction methods, this paper proposes a novel prediction method for building energy consumption, which is based on the seamless integration of the deep neural network and transfer reinforcement learning (DNN-TRL). The method introduces a stack denoising autoencoder to extract the deep features of the building energy consumption, and shares the hidden layer structure to transfer the common information between different building energy consumption problems. The output of the DNN model is used as the input of the Sarsa algorithm to improve the prediction performance of the target building energy consumption. To verify the performance of the DNN-TRL algorithm, based on the data recorded by American Power Balti Gas and Electric Power Company, and compared with Sarsa, ADE-BPNN, and BP-Adaboost algorithms, the experimental results show that the DNN-TRL algorithm can effectively improve the prediction accuracy of the building energy consumption.

关键词： DNN-TRL feature transfer denoising autoencoder building energy prediction

来源：评论

学校读者我要写书评

暂无评论

Enhanced recognition of bonding interface defects within CFRP-steel adhesive structures via thermal signals in low-power vibrothermography

引用

CONSTRUCTION AND BUILDING MATERIALS 2024年 456卷

作者： Zhang, Yubin Xu, Changhang Liu, Pengqian Xie, Jing Liu, Rui Zhao, Qing China Univ Petr East China Coll Mech & Elect Engn Qingdao 266580 Peoples R China China Univ Petr East China State Key Lab Chem Safety Qingdao 266580 Peoples R China

Carbon fiber reinforced polymers (CFRP) have been used as one of the options to strengthen steel structures through adhesive bonding, particularly in specific applications where traditional strengthening methods may not be suitable. Therefore, it becomes crucial to perform inspections on the resulting CFRP-steel adhesive structures (CSAS) to ensure their structural integrity and safety. However, the distinct physical properties of CFRP, epoxy resin, and steel pose significant challenges to accurately inspecting bonding interface defects of such special hybrid engineering structures. To address these challenges, a new approach, streamlined one-dimensional convolutional denoising autoencoder-low-power vibrothermography (SOCDAE-LVT), is proposed in this study to enhance the recognition of bonding interface defects within CSAS. This approach utilizes thermal signals from low-power vibrothermography (LVT) to enhance the recognizability of CSAS bonding interface defects. A lowpower vibrothermography inspection system was developed to acquire thermal signals on the surface of CSAS samples. A streamlined one-dimensional convolutional denoising autoencoder (SOCDAE) model was designed for robust representation extraction of the thermal signal at each pixel point. The study further investigated the impact of different types of added noise and signal pre-processing approaches on the performance of the SOCDAE-LVT, aiming to optimize its effectiveness. By comparing qualitatively and quantitatively with the stateof-the-art approaches, the results show that the proposed approach can better improve the recognizability of defects. The enhanced recognizability of bonding interface defects enables accurate assessment of the quality of CSAS, thereby contributing to the safety of such structures.

关键词： CFRP-steel adhesive structures Bonding interface defects Low-power vibrothermography Thermal signals Streamlined one-dimensional convolutional denoising autoencoder

来源：评论

学校读者我要写书评

暂无评论

SMPLR: Deep learning based SMPL reverse for 3D human pose and shape recovery

引用

PATTERN RECOGNITION 2020年 106卷 107472-107472页

作者： Madadi, Meysam Bertiche, Hugo Escalera, Sergio Comp Vis Ctr Edif OCampus UAB Catalonia 08193 Barcelona Spain Univ Barcelona Dept Math & Informat Gran Via Corts Catalanes Barcelona 58508007 Spain

In this paper we propose to embed SMPL within a deep-based model to accurately estimate 3D pose and shape from a still RGB image. We use CNN-based 3D joint predictions as an intermediate representation to regress SMPL pose and shape parameters. Later, 3D joints are reconstructed again in the SMPL output. This module can be seen as an autoencoder where the encoder is a deep neural network and the decoder is SMPL model. We refer to this as SMPL reverse (SMPLR). By implementing SMPLR as an encoder-decoder we avoid the need of complex constraints on pose and shape. Furthermore, given that in-the-wild datasets usually lack accurate 3D annotations, it is desirable to lift 2D joints to 3D without pairing 3D annotations with RGB images. Therefore, we also propose a denoising autoencoder (DAE) module between CNN and SMPLR, able to lift 2D joints to 3D and partially recover from structured error. We evaluate our method on SURREAL and Human3.6M datasets, showing improvement over SMPL-based state-of-the-art alternatives by about 4 and 12 mm, respectively. (C) 2020 Elsevier Ltd. All rights reserved.

关键词： Deep learning 3D Human pose Body shape SMPL denoising autoencoder Volumetric stack hourglass

来源：评论

学校读者我要写书评

暂无评论

Hybrid unsupervised web-attack detection and classification-A deep learning approach

引用

COMPUTER STANDARDS & INTERFACES 2023年 86卷

作者： Pillai, Seema Sharma, Anurag MATS Univ Arang 493441 Chhattisgarh India MATS Univ MATS Sch Engn Comp Sci Engn Arang India

Web requests made by users of web applications are manipulated by hackers to gain control of web servers. Moreover, detecting web attacks has been increasingly important in the distribution of information over the last few decades. Also, several existing techniques had been performed on detecting vulnerable web attacks using machine learning and deep learning techniques. However, there is a lack in achieving attack detection ratio owing to the utilization of supervised and semi-supervised learning approaches. Thus to overcome the afore-mentioned issues, this research proposes a hybrid unsupervised detection model a deep learning-based anomaly -based web attack detection. Whereas, the encoded outputs of De-Noising autoencoder (DAE), as well as Stacked autoencoder (SAE), are integrated and given to the Generative adversarial network (GAN) as input to improve the feature representation ability to detect the web attacks. Consequently, for classifying the type of attacks, a novel DBM-Bi LSTM-based classification model has been introduced. Which incorporates DBM for binary clas-sification and Bi-LSTM for multi-class classification to classify the various attacks. Finally, the performance of the classifier in terms of recall, precision, F1-Score, and accuracy are evaluated and compared. The proposed method achieved high accuracy of 98%.

关键词： SQL injection attack Cross -site scripting attacks denoising autoencoder Deep Boltzmann machine Binary long term short term memory

来源：评论

学校读者我要写书评

暂无评论

Application of stack marginalised sparse denoising auto-encoder in fault diagnosis of rolling bearing

引用

JOURNAL OF ENGINEERING-JOE 2018年第16期2018卷 1772-1777页

作者： Zhang, Junling Chen, Zhigang Du, Xiaolei Xu, Xu Yu, Miao Beijing Univ Civil Engn & Architecture Inst Elect & Mech & Vehicle Engn Beijing 100044 Peoples R China Construct Safety Monitoring Engn Technol Res Ctr Beijing 100037 Peoples R China

When a fracturing vehicle is working, it generally needs to bear high loads, media corrosion and erosion. For this special working environment, this study proposes a rolling bearing fault diagnosis method based on stack marginalised sparse denoising auto-encoder (SDAE). This method combines the sparse auto-encoder (SAE) and the denoising auto-encoder (DAE) and combines the characteristics of dimensionality reduction and robustness. The method adds marginalisation to optimise the SDAE. Finally, it uses a two-layer stacking method. The output results of the second marginalised SDAE are used as input to the softmax classifier for learning training and classification testing. This improved method (stack SDAE) improves the denoising ability, reduces the computational complexity, solves the problems of difficult parameter adjustment and slows training convergence. The experimental tests were carried out on the failure of pitting corrosion of the outer ring of the bearing, pitting failure of the inner ring, and cracking of the rolling element. The results show that the algorithm can effectively improve the accuracy of fault diagnosis of rolling bearings, and it has greatly improved than the algorithms of SAEs and DAE.

关键词： rolling bearings fault diagnosis mechanical engineering computing computational complexity corrosion failure analysis cracks fracture signal denoising media corrosion erosion rolling bearing fault diagnosis method two-layer stacking method SDAE stack marginalised sparse denoising autoencoder application fracturing vehicle sparse autoencoder SAE denoising autoencoder DAE softmax classifier learning training classification testing computational complexity pitting corrosion failure cracking

来源：评论

学校读者我要写书评

暂无评论

Deep transfer learning for rolling bearing fault diagnosis under variable operating conditions

引用

ADVANCES IN MECHANICAL ENGINEERING 2019年第12期11卷

作者： Che, Changchang Wang, Huawei Fu, Qiang Ni, Xiaomei Nanjing Univ Aeronaut & Astronaut Coll Civil Aviat 29 Jiang Jun St Nanjing 211106 Jiangsu Peoples R China

Rolling bearings are the vital components of rotary machines. The collected data of rolling bearing have strong noise interference, massive unlabeled samples, and different fault features. Thus, a deep transfer learning method is proposed for rolling bearings fault diagnosis under variable operating conditions. To obtain robust feature representation, the denoising autoencoder is used to denoise and reduce dimension of unlabeled rolling bearing signals. For those unlabeled target domain signals, a feature matching method based on multi-kernel maximum mean discrepancies between source domain and target domain is adopted to get enough labeled target domain samples. Then, these rolling bearing signals are converted to multi-dimensional graph samples and fed into a convolutional neural network model for fault diagnosis. To improve the generalization of convolutional neural network under variable operating conditions, we combine model-based transfer learning with feature-based transfer learning to initialize and optimize the convolutional neural network parameters. The effectiveness of the proposed method is validated through several comparative experiments of Case Western Reserve University data. The results demonstrate that the proposed method can learn features adaptively from noisy data and increase the accuracy rate by 2%-8% comparing with other models.

关键词： denoising autoencoder convolutional neural network multi-kernel maximum mean discrepancy transfer learning fault diagnosis rolling bearing

来源：评论

学校读者我要写书评

暂无评论

Biomedical ontology alignment: an approach based on representation learning

引用

JOURNAL OF BIOMEDICAL SEMANTICS 2018年第1期9卷 21页

作者： Kolyvakis, Prodromos Kalousis, Alexandros Smith, Barry Kiritsis, Dimitris Ecole Polytech Fed Lausanne Route Cantonale CH-1015 Lausanne Switzerland Univ Appl Sci HES SO Western Switzerland Carouge Business Informat Dept Lausanne Switzerland SUNY Buffalo Dept Philosophy 104 Pk Hall Buffalo NY 14260 USA SUNY Buffalo Dept Biomed Informat 104 Pk Hall Buffalo NY 14260 USA

Background: While representation learning techniques have shown great promise in application to a number of different NLP tasks, they have had little impact on the problem of ontology matching. Unlike past work that has focused on feature engineering, we present a novel representation learning approach that is tailored to the ontology matching task. Our approach is based on embedding ontological terms in a high-dimensional Euclidean space. This embedding is derived on the basis of a novel phrase retrofitting strategy through which semantic similarity information becomes inscribed onto fields of pre-trained word vectors. The resulting framework also incorporates a novel outlier detection mechanism based on a denoising autoencoder that is shown to improve performance. Results: An ontology matching system derived using the proposed framework achieved an F-score of 94% on an alignment scenario involving the Adult Mouse Anatomical Dictionary and the Foundational Model of Anatomy ontology (FMA) as targets. This compares favorably with the best performing systems on the Ontology Alignment Evaluation Initiative anatomy challenge. We performed additional experiments on aligning FMA to NCI Thesaurus and to SNOMED CT based on a reference alignment extracted from the UMLS Metathesaurus. Our system obtained overall F-scores of 93.2% and 89.2% for these experiments, thus achieving state-of-the-art results. Conclusions: Our proposed representation learning approach leverages terminological embeddings to capture semantic similarity. Our results provide evidence that the approach produces embeddings that are especially well tailored to the ontology matching task, demonstrating a novel pathway for the problem.

关键词： Ontology matching Semantic similarity Sentence embeddings Word embeddings denoising autoencoder Outlier detection

来源：评论

学校读者我要写书评

暂无评论

Deep feature fusion with computer vision driven fall detection approach for enhanced assisted living safety

引用

SCIENTIFIC REPORTS 2024年第1期14卷 1-17页

作者： Almukadi, Wafa Sulaiman Alrowais, Fadwa Saeed, Muhammad Kashif Yahya, Abdulsamad Ebrahim Mahmud, Ahmed Marzouk, Radwa Univ Jeddah Coll Comp Sci & Engn Dept Software Engn Jeddah Saudi Arabia Princess Nourah Bint Abdulrahman Univ PNU Coll Comp & Informat Sci Comp Sci Dept POB 84428 Riyadh 11671 Saudi Arabia King Khalid Univ Appl Coll Mahayil Dept Comp Sci Abha Saudi Arabia Univ Northern Border Coll Comp & Informat Technol Informat Technol Dept Rafha Saudi Arabia Future Univ Egypt Res Ctr New Cairo 11835 Egypt Princess Nourah Bint Abdulrahman Univ Coll Comp & Informat Sci Dept Informat Syst Riyadh Saudi Arabia

Assisted living facilities cater to the demands of the elderly population, providing assistance and support with day-to-day activities. Fall detection is fundamental to ensuring their well-being and safety. Falls are frequent among older persons and might cause severe injuries and complications. Incorporating computer vision techniques into assisted living environments is revolutionary for these issues. By leveraging cameras and complicated approaches, a computer vision (CV) system can monitor residents' movements continuously and identify any potential fall events in real time. CV, driven by deep learning (DL) techniques, allows continuous surveillance of people through cameras, investigating complicated visual information to detect potential fall risks or any instances of falls quickly. This system can learn from many visual data by leveraging DL, improving its capability to identify falls while minimalizing false alarms precisely. Incorporating CV and DL enhances the efficiency and reliability of fall detection and allows proactive intervention, considerably decreasing response times in emergencies. This study introduces a new Deep Feature Fusion with Computer Vision for Fall Detection and Classification (DFFCV-FDC) technique. The primary purpose of the DFFCV-FDC approach is to employ the CV concept for detecting fall events. Accordingly, the DFFCV-FDC approach uses the Gaussian filtering (GF) approach for noise eradication. Besides, a deep feature fusion process comprising MobileNet, DenseNet, and ResNet models is involved. To improve the performance of the DFFCV-FDC technique, improved pelican optimization algorithm (IPOA) based hyperparameter selection is performed. Finally, the detection of falls is identified using the denoising autoencoder (DAE) model. The performance analysis of the DFFCV-FDC methodology was examined on the benchmark fall database. A widespread comparative study reported the supremacy of the DFFCV-FDC approach with existing techniques.

关键词： Fall detection Computer vision Deep feature fusion Pelican optimization algorithm Deep learning denoising autoencoder

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

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

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

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

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：