检索结果-内蒙古大学图书馆

Deep learning-based integrated attack detection framework to protect distance relays against cyberattacks

ELECTRIC POWER SYSTEMS RESEARCH 2024年 231卷

作者： Narang, Jagendra Kumar Bag, Baidyanath Natl Inst Technol Raipur Dept Elect Engn GE Rd Raipur 492010 Chhattisgarh India

The cyber security of communication -assisted intelligent relays at the IEC61850-enabled modern digital substations has become a prime concern among power companies and regulatory bodies. Particularly, distance relay IEDs are easy to manipulate to send false trip command to circuit breakers through random false data injection and replay attacks. With these attacks evolving and becoming more sophisticated, existing intrusion detection techniques can be easily compromised, resulting in unwanted line outages. This work proposes a convolutional -based autoencoder and Siamese neural network deep learning integrated attack detection framework (IADF) to differentiate these attacks from actual faults. The framework works in conjunction with distance relay IEDs to issue genuine trip commands to circuit breakers. The autoencoder, trained on fault voltage and current signals, detects the false data injection attacks using reconstruction error, while the siamese neural network trained on similar and dissimilar pairs of the fault voltage and current signals detects the fully synchronized replay attacks using similarity estimation with fault data stored on database. The validity of the proposed framework is assessed using IEEE 39 and 118 bus test system simulated on PSCAD/EMTDC software. The validation results suggest that the proposed IADF can accurately discriminate the faults from attacks and is resilient to various system parameter changes.

关键词： False data injection attack on distance relays Replay attacks False tripping command Deep learning detection of cyber attacks convolutional autoencoder convolutional siamese neural network

来源：评论

学校读者我要写书评

暂无评论

Interpretable regional meteorological feature extraction enhances deep learning for extended 120-h PM2.5 forecasting

引用

JOURNAL OF CLEANER PRODUCTION 2024年 483卷

作者： Liu, Xinyi Pu, Xueting Lu, Chengwei Zhang, Han Li, Tao Grieneisen, Michael L. Li, Jucheng Ma, Ning Yan, Chang Zhan, Yu Yang, Fumo Sichuan Univ Dept Environm Sci & Engn Chengdu 610065 Peoples R China Sichuan Univ Coll Carbon Neutral Future Technol Chengdu 610065 Peoples R China Chengdu Acad Environm Sci Chengdu 610072 Peoples R China State Grid Sichuan Elect Power Res Inst Chengdu 610041 Peoples R China Univ Calif Davis Dept Land Air & Water Resources Davis CA 95616 USA

While deep learning models perform well in short-term PM2.5 forecasting, their performance tends to decay significantly with increasing forecast spans. This study proposes a "Domain-to-Point" approach to enhance deep learning models for extended 120-h PM2.5 forecasting. We used convolutional autoencoders (ConvAEs) to extract regional meteorological features (RMFs) from both surrounding and future meteorological data provided by an advanced weather forecasting system. The RMFs and temporal indicators were used as predictor variables in a time-attention-based bidirectional long short-term memory neural network to forecast the next 24- to 120-h PM2.5. For four megacities in China, the ConvAEs compressed the high-dimensional meteorological data into a lower-dimensional representation, while preserving most of the relevant information (R-2 > 0.75). Using the RMFs as inputs substantially mitigated the performance decay of the PM2.5 forecasting models over extended forecast spans, enabling accurate longer-term forecasting, with R-2 improvements of 14.7-79.2% for the 120-h forecasts. This study further revealed that errors in the meteorological forecasts are the primary cause of performance decay, suggesting that improving the accuracy of meteorological forecasts could be an effective approach for enhancing PM2.5 forecasting performance. With Beijing as a case study, model interpretation using the Shapley-additive-explanations method demonstrated the crucial role played by RMFs in accurately forecasting a typical air pollution episode and identifying the main meteorological factors affecting pollution. The proposed "Domain-to-Point" approach enhances the capability of deep learning models for longer-term PM2.5 forecasting, thus facilitating the development of preventative measures to safeguard public health against air pollution.

关键词： PM2.5 forecasting External feature engineering Meteorological feature extraction Deep learning convolutional autoencoder

来源：评论

学校读者我要写书评

暂无评论

A novel self-attention deep subspace clustering

引用

INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS 2021年第8期12卷 2377-2387页

作者： Chen, Zhengfan Ding, Shifei Hou, Haiwei China Univ Min & Technol Sch Comp Sci & Technol Xuzhou 221116 Jiangsu Peoples R China Minist Educ People S Republ China Mine Digitizat Engn Res Ctr Xuzhou 221116 Jiangsu Peoples R China

Most of the existing deep subspace clustering methods leverage convolutional autoencoders to obtain feature representation for non-linear data points. These methods commonly adopt the structure of a few convolutional layers because stacking many convolutional layers may cause computationally inefficient and optimization difficulties. However, long-range dependencies can hardly be captured when convolutional operations are not repeated enough, thus affect the quality of feature extraction which the performance of deep subspace clustering method highly lies in. To deal with this issue, we propose a novel self-attention deep subspace clustering (SADSC) model, which learns more favorable data representations by introducing self-attention mechanisms into convolutional autoencoders. Specifically, SADSC leverages three convolutional layers and add the self-attention layers after the first and third ones in encoders, then decoders have symmetric structures. The self-attention layers maintain the variable input sizes and can be easily combined with different convolutional layers in autoencoder. Experimental results on the handwritten recognition, face and object clustering datasets demonstrate the advantages of SADSC over the state-of-the-art deep subspace clustering models.

关键词： Deep subspace clustering convolutional autoencoder Self-attention

来源：评论

学校读者我要写书评

暂无评论

Multi-objective Genetic Algorithm Based Deep Learning Model for Automated COVID-19 Detection Using Medical Image Data

引用

JOURNAL OF MEDICAL AND BIOLOGICAL ENGINEERING 2021年第5期41卷 678-689页

作者： Bansal, S. Singh, M. Dubey, R. K. Panigrahi, B. K. Indian Inst Technol Delhi Comp Sci & Engn Dept New Delhi 110016 India Robert Bosch Engn & Business Solut Private Ltd He 123 Hosur Rd7th Block Bengaluru 560095 Karnataka India Indian Inst Technol Delhi Elect Engn Dept New Delhi 110016 India

Purpose In early 2020, the world is amid a significant pandemic due to the novel coronavirus disease outbreak, commonly called the COVID-19. Coronavirus is a lung infection disease caused by the Severe Acute Respiratory Syndrome Coronavirus 2 virus (SARS-CoV-2). Because of its high transmission rate, it is crucial to detect cases as soon as possible to effectively control the spread of this pandemic and treat patients in the early stages. RT-PCR-based kits are the current standard kits used for COVID-19 diagnosis, but these tests take much time despite their high precision. A faster automated diagnostic tool is required for the effective screening of COVID-19. Methods In this study, a new semi-supervised feature learning technique is proposed to screen COVID-19 patients using chest CT scans. The model proposed in this study uses a three-step architecture, consisting of a convolutional autoencoder based unsupervised feature extractor, a multi-objective genetic algorithm (MOGA) based feature selector, and a Bagging Ensemble of support vector machines based binary classifier. The proposed architecture has been designed to provide precise and robust diagnostics for binary classification (COVID ***). A dataset of 1252 COVID-19 CT scan images, collected from 60 patients, has been used to train and evaluate the model. Results The best performing classifier within 127 ms per image achieved an accuracy of 98.79%, the precision of 98.47%, area under curve of 0.998, and an F1 score of 98.85% on 497 test images. The proposed model outperforms the current state of the art COVID-19 diagnostic techniques in terms of speed and accuracy. Conclusion The experimental results prove the superiority of the proposed methodology in comparison to existing *** study also comprehensively compares various feature selection techniques and highlights the importance of feature selection in medical image data problems.

关键词： Coronavirus (COVID-19) convolutional autoencoder Multi-objective genetic algorithm Feature subset selection

来源：评论

学校读者我要写书评

暂无评论

Semi-supervised learning for laser directed energy deposition monitoring via co-axial dynamic imaging

引用

ADDITIVE MANUFACTURING 2025年 97卷

作者： Zheng, Fenglei Xie, Luofeng Bai, Qingsong Zhu, Yangyang Yin, Ming Zhang, Yuhang Niu, Kaiyu Sichuan Univ Sch Mech & Engn 24 South Sect 1Yihuan Rd Chengdu 610065 Peoples R China

Data-driven approaches have emerged as a promising solution for monitoring the quality of metal parts produced by Laser Directed Energy Deposition (LDED). Current methods often require specialized knowledge for extensive data labeling, while defect characterization experiments can be costly and time-consuming. To overcome these issues, we propose a semi-supervised autoencoder framework capable of real-time prediction of local quality attributes during the LDED process. Our method significantly diverges from traditional semi-supervised learning approaches, which primarily rely on data augmentation to enforce consistency regularization. First, domain-level samples are constructed based on a series of coaxial melt pool images, providing a comprehensive assessment of the quality state of local regions by analyzing the local porosity of the samples. Second, a semi-supervised convolutional autoencoder (SCAE) is designed to develop a robust understanding of melt pool features and latent knowledge through unsupervised training. Residual connections and L2 normalization are utilized to enhance the extraction of melt pool morphological features and effectively handle the multi-scale nature of pore samples, respectively. Finally, a limited amount of labeled data is employed to activate the class space of the samples. Experimental results validate the feasibility and efficacy of the proposed model in monitoring the local quality of the LDED process. With only 60 % of the original labeled data, our model's accuracy is comparable to the maximum accuracy achieved by the fully supervised model. Even with a reduction of labeled data to 20 %, our model still maintains a prediction accuracy of 83.8 %, significantly reducing the time and costs associated with data labeling. These findings contribute to improving the quality of LDED final products and promote its broader industrial application.

关键词： Laser directed energy deposition Semi-supervised Real-time monitoring convolutional autoencoder

来源：评论

学校读者我要写书评

暂无评论

A Framework for Anomaly Classification and Segmentation in Remanufacturing using autoencoders and Simulated Data

引用

Procedia CIRP 2024年 126卷 1005-1010页

作者： Jan-Philipp Kaiser Carl-Leandro Enslin Erik Tabuchi Barczak Florian Stamer Michael Heizmann Gisela Lanza wbk Institute of Production Science Karlsruhe Institute of Technology (KIT) Kaiserstr. 12 76131 Karlsruhe IIIT – Institute of Industrial Information Technology Karlsruhe Institute of Technology (KIT) Hertzstr. 16 76187 Karlsruhe Global Advanced Manufacturing Institute (GAMI) KIT China Branch 215123 Suzhou China

In remanufacturing, reusable end-of-life products must be separated from those for which remanufacturing is not economical. In this process, the focus is initially on the detection and evaluation of surface defects of the used product. Due to the use phase of the product, these show a great variety with strongly differing degrees of severity. Deep-learning-based autoencoders offer the advantage of defect-independent detection of anomalies by detecting deviations from a defect-free product condition. In the present work, anomaly detection based on synthetically generated image data is investigated. In addition, a framework for classification and segmentation is developed and presented. The results show that the approach presented in the paper is effective in detecting anomalies with potential for improvement through adjustments such as perceptual loss or image processing methods. However, challenges remain in accurately detecting anomalies in the presence of missing parts, large anomalies, and image data with high variance.

关键词： Remanufacturing Anomaly Detection Unsupervised Learning convolutional autoencoder

来源：评论

学校读者我要写书评

暂无评论

Improvement of impact acoustic inspection for high-speed railway track slabs using time-frequency analysis and non-defective machine learning

引用

NDT & E INTERNATIONAL 2024年 145卷

作者： Inaba, Kohko Tanigawa, Hikaru Naito, Hideki Railway Tech Res Inst 2-8-38 Hikari ChoKokubunji Tokyo 1858540 Japan Tohoku Univ 6-6-06Aramaki Aza AobaAoba Ku Sendai Miyagi 9808579 Japan

Slab track is one of the most highly used ballastless track forms in Japan ' s high-speed railway system ( Shin- kansen ) and is composed from track slabs (precast RC or PRC member), a filling layer (mixture of cement, asphalt emulsion and fine aggregates) and concrete bed. As they age, it has been reported that the supporting conditions of track slabs change due to damage in the filling layer that in turn affects running stability. It is necessary to evaluate supporting conditions of track slabs accurately during maintenance of slab tracks prior to signs of cracking. Some previous works show the usefulness of impact acoustics and non -defective machine learning using frequency response spectrums as feature values for evaluating supporting conditions of track slabs. In this study, we examine a method of non -defective learning to improve the evaluation of supporting conditions: timefrequency response characteristics of impact acoustics using time -frequency analysis and resulting spectrograms as feature values. As a result, we have improved accuracy rate (from 81.90 % to 86.72 %) and F -value (from 70.79 % to 78.75 %) more precisely than conventional method based on frequency response functions as feature values.

关键词： Slab track Impact acoustic inspection Continuous wavelet transform Principal component analysis convolutional autoencoder

来源：评论

学校读者我要写书评

暂无评论

An elevator door anomaly detection method based on improved deep multi-sphere support vector data description

引用

COMPUTERS & ELECTRICAL ENGINEERING 2024年 120卷

作者： Xie, Pengdong Zhang, Linxuan Li, Minghong Qiu, Chaojie Tsinghua Univ Shenzhen Int Grad Sch Shenzhen 518057 Peoples R China Tsinghua Univ Dept Automat Beijing 100084 Peoples R China Xinjiang Univ Sch Elect Engn Urumqi 830017 Peoples R China

Various types of elevator door faults and difficulties in fault data acquisition make it difficult to use supervised learning methods for fault diagnosis. This paper proposes a semi-supervised anomaly detection method based on improved deep multi-sphere support vector data description. Multiple distinguishing hyper-spheres, characterized by minimum volume, are established on the foundation of normal data by this method. These hyper-spheres represent the multi- modal distribution exhibited by the normal data. In addition, the method fuses multi-sensor source data such as tri-axial acceleration, dual-axial tilt angle, and introduces the structure of InceptionTime to realize the fusion of multivariate data and feature extraction in multiple resolutions. Experiments verify the feasibility of the method with an overall AUC of 96.50%, and comparative experiments demonstrate the superior detection performance. This contributes a novel, accurate, and more appropriate method to the elevator door anomaly detection.

关键词： Anomaly detection Support vector data description Semi-supervised learning convolutional autoencoder

来源：评论

学校读者我要写书评

暂无评论

Machine learning-based prediction of swirl combustor operation from flame imaging

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2024年 138卷

作者： Bong, Cheolwoo Ali, Mohammed H. A. Im, Seong kyun Do, Hyungrok Bak, Moon Soo Sungkyunkwan Univ Sch Mech Engn Suwon 16419 South Korea Korea Univ Dept Mech Engn Seoul 02841 South Korea Seoul Natl Univ Dept Mech Engn Seoul 08826 South Korea Sungkyunkwan Univ Dept Smart Fab Technol Suwon 16419 South Korea

This study proposes a novel data-driven model to distinguish normal operation of a swirl combustor and predict key operation conditions using a flame image taken with a low-cost monochrome camera. The model, in the form of a convolutional neural network (CNN), is designed to take a flame image as input and provide either the total air flow rate (Q) or the fuel-air equivalence ratio (phi) as an output. However, since the type of problem in this study is regression, it is necessary to make predictions only on normal operation images, as it is not feasible to collect flame images for all unstable combustion modes. Thus, the stacked convolutional layers were first trained as a convolutional autoencoder (CAE) in an unsupervised manner using only flame images under normal operation modes, so that the CAE can perform well only on normal operation images. Then, a regressor that outputs either Q or phi is connected to the trained encoder and trained in a supervised manner. It was found that the model can predict Q and phi within +5.17 L/min (equivalent to 3.4% of the total flow rate) and +0.026, respectively, with a 96% probability, along with detecting abnormalities based on large reconstruction errors of input images. Predictions and image collection can be performed within 50 ms, demonstrating the potential for real-time monitoring of combustor status.

关键词： Swirl combustor Operation condition monitoring Abnormality detection convolutional autoencoder Gradient-weighted activation mapping

来源：评论

学校读者我要写书评

暂无评论

Fault Diagnosis in Centrifugal Pumps: A Dual-Scalogram Approach with Convolution autoencoder and Artificial Neural Network

引用

SENSORS 2024年第3期24卷 851页

作者： Zaman, Wasim Ahmad, Zahoor Kim, Jong-Myon Univ Ulsan Dept Elect Elect & Comp Engn Ulsan 44610 South Korea Prognosis & Diagnost Technol Co Ltd Ulsan 44610 South Korea

This paper proposes a new fault diagnosis method for centrifugal pumps by combining signal processing with deep learning techniques. Centrifugal pumps facilitate fluid transport through the energy generated by the impeller. Throughout the operation, variations in the fluid pressure at the pump's inlet may impact the generalization of traditional machine learning models trained on raw statistical features. To address this concern, first, vibration signals are collected from centrifugal pumps, followed by the application of a lowpass filter to isolate frequencies indicative of faults. These signals are then subjected to a continuous wavelet transform and Stockwell transform, generating two distinct time-frequency scalograms. The Sobel filter is employed to further highlight essential features within these scalograms. For feature extraction, this approach employs two parallel convolutional autoencoders, each tailored for a specific scalogram type. Subsequently, extracted features are merged into a unified feature pool, which forms the basis for training a two-layer artificial neural network, with the aim of achieving accurate fault classification. The proposed method is validated using three distinct datasets obtained from the centrifugal pump under varying inlet fluid pressures. The results demonstrate classification accuracies of 100%, 99.2%, and 98.8% for each dataset, surpassing the accuracies achieved by the reference comparison methods.

关键词： centrifugal pump Stockwell transform fault diagnosis rotating machinery convolutional autoencoder vibrational signals deep learning

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：