检索结果-内蒙古大学图书馆

Deep Learning-Based Residual Useful Lifetime Prediction for Assets With Uncertain Failure Modes

Journal of Computing and Information Science in engineering 2025年第4期25卷 041002页

作者： Su, Yuqi Fang, Xiaolei Operations Research Graduate Program North Carolina State University 915 Partners Way Raleigh NC 27607 Edward P. Fitts Department of Industrial and Systems Engineering North Carolina State University 915 Partners Way Raleigh NC 27607

industrial prognostics focuses on utilizing degradation signals to forecast and continually update the residual useful life of complex engineering systems. However, existing prognostic models for systems with multiple failure modes face several challenges in real-world applications, including overlapping degradation signals from multiple components, the presence of unlabeled historical data, and the similarity of signals across different failure modes. To tackle these issues, this research introduces two prognostic models that integrate the mixture (log)-location-scale distribution with deep learning. This integration facilitates the modeling of overlapping degradation signals, eliminates the need for explicit failure mode identification, and utilizes deep learning to capture complex nonlinear relationships between degradation signals and residual useful lifetimes. Numerical studies validate the superior performance of these proposed models compared to existing methods.

关键词： Failure mechanisms mixture (log)-location-scale distribution Failure Weibull distribution multiple failure modes Signals Service life (Equipment) prognostics data analytics Deep learning machine learning for engineering applications engineering systems and industry applications industrial Internet of things big data and analytics

来源：评论

学校读者我要写书评

暂无评论

Differentially Private Log-Location-Scale Regression Using Functional Mechanism

arXiv

引用

arXiv 2024年

作者： Sheng, Jiewen Fang, Xiaolei Graduate Program in Operations Research North Carolina State University United States Edward P. Fitts Department of Industrial and Systems Engineering North Carolina State University United States

This article introduces differentially private log-location-scale (DP-LLS) regression models, which incorporate differential privacy into LLS regression through the functional mechanism. The proposed models are established by injecting noise into the log-likelihood function of LLS regression for perturbed parameter estimation. We will derive the sensitivities utilized to determine the magnitude of the injected noise and prove that the proposed DP-LLS models satisfy ϵ-differential privacy. In addition, we will conduct simulations and case studies to evaluate the performance of the proposed models. The findings suggest that predictor dimension, training sample size, and privacy budget are three key factors impacting the performance of the proposed DP-LLS regression models. Moreover, the results indicate that a sufficiently large training dataset is needed to simultaneously ensure decent performance of the proposed models and achieve a satisfactory level of privacy protection. © 2024, CC BY.

关键词： Budget control

来源：评论

学校读者我要写书评

暂无评论

A Method to Gaze Following Detection by Computer Vision Applied to Production Environments 17th

A Method to Gaze Following Detection by Computer Vision Appl...

引用

17th IFIP WG 5.1 International Conference on Product Lifecycle Management, PLM 2020

作者： Cenzi, Emannuell Dartora Rudek, Marcelo Pontifícia Universidade Católica do Paraná PUCPR Industrial and Systems Engineering Graduate Program – PPGEPS CuritibaPR80215-901 Brazil

ISBN: (纸本)9783030628062

The humans have the natural ability of following objects with the head and eyes and identify the relationship between those objects. This daily activity represents a challenge for computer vision systems. The procedure to identify the relationship between human eye gaze and the trackable objects is complex and demands several details. In this current paper we proposed a review of the main gazing following methods, identified the respective performance of them and also proposed an AI based method to estimate the gaze from 2D images based on head pose estimation. The main important details to be recovered from images are scene depth, head position and alignment and ocular rotation. In this approach we perform a track estimation of the gaze direction without the use of the eye position, and also, the face partial occlusion is considered in the analysis. The proposed approach allows low cost in processing with considerable accuracy at low complexity sceneries, because we don’t need to extract the facial features. Gaze tracking is important to evaluate employees’ attention to specific tasks in order to prevent accidents and improve work quality. The presented method improves the current knowing workflow by applying the head pose estimation instead face detection for training and inference. The promisors results are presented and open points are also discussed. © 2020, IFIP International Federation for Information Processing.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Virtual Wearable Sensor Data Generation with Generative Adversarial Networks

Virtual Wearable Sensor Data Generation with Generative Adve...

引用

Simulation Winter Conference

作者： Yining Huang Hong Wan Xi Chen Operations Research Graduate Program North Carolina State University Raleigh NC USA Department of Industrial and Systems Engineering North Carolina State University Raleigh NC USA Grado Department of Industrial and Systems Engineering Virginia Tech Blacksburg VA USA

This study delves into the utilization of Generative Adversarial Networks (GANs) for generating subject-specific time series sensor data, offeringaninnovativealternativetotraditionalmetamodel-basedsimulations. We undertake an in-depth analysis of DoppelGANger, a prominent GAN variant for time series data and metadata generation, evaluating its efficiency and efficacy. The sensor data for this investigation was sourced from the National Health and Nutrition Examination Survey, which served as the foundational training set. We scrutinized the synthesized sensor data corresponding to various physical attributes, focusing on the temporal and multi-dimensional statistical properties. Our empirical findings underscore the potential of GANs to adeptly capture the time-dependent correlations and the intricate statistical characteristics inherent in multi-dimensional data. This insight into GANs’ capabilities is a crucial step towards more sophisticated synthetic data generation, with significant implications for future applications in wearable technology and personalized health monitoring systems.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Generative assistant for digital twin simulations

引用

Procedia CIRP 2025年 132卷 129-134页

作者： Pedro Antonio Boareto Eduardo de Freitas Rocha Loures Eduardo Alves Portela Santos Fernando Deschamps Industrial and Systems Engineering Graduate Program Pontifical Catholic University of Paraná 1155 Imaculada Conceição St. Curitiba Paraná 80215-901 Brazil Federal University of Paraná 632 Prefeito Lothário Meissner Av. Curitiba Paraná 80210-170 Brazil

One of the key emerging technologies in Industry 4.0 is the Digital Twin (DT). Although it promises increased efficiency, productivity, and innovation, its adoption faces challenges such as high investment costs and the need for workforce requalification. Generative Artificial Intelligence (GAI) emerges as a promising solution, offering capabilities to accelerate development processes and reduce costs. This study aims to leverage GAI to enhance the development of DT and support decision-making in industrial environments by proposing a Generative Assistant for Digital Twin Simulations (GADTS). This proposal generates operational models quickly, offers greater customization, and facilitates the creation of efficient scenario simulations in natural language. The proposal was tested with artificial data. As a result, the development of highly personalized DT simulations with Key Performance Indicators (KPIs) was entirely abstracted into natural language requests.

关键词： Process Mining Discrete Event Simulation Generative Artificial Intelligence Digital Twin Industry 4.0 Chatbot Decision-making

来源：评论

学校读者我要写书评

暂无评论

Multiband Patch Antenna Design Using Nature-Inspired Optimization Method

IEEE Open Journal of Antennas and Propagation

引用

IEEE Open Journal of Antennas and Propagation 2021年 2卷 151-162页

作者： Boursianis, Achilles D. Papadopoulou, Maria S. Pierezan, Juliano Mariani, Viviana C. Coelho, Leandro S. Sarigiannidis, Panagiotis Koulouridis, Stavros Goudos, Sotirios K. ELEDIA@AUTH Research Group School of Physics Aristotle University of Thessaloniki Thessaloniki54124 Greece Department of Electrical Engineering Federal University of Parana Curitiba80060-000 Brazil Industrial and Systems Engineering Graduate Program and Mechanical Engineering Graduate Program Pontifical Catholic University of Parana Curitiba80215-901 Brazil Department of Electrical and Computer Engineering University of Western Macedonia Kozani501 00 Greece Electrical and Computer Engineering Department University of Patras Patras26504 Greece

Radio frequency energy harvesting has attracted considerable interest as a technique of enabling self-powered wireless networks. This technique faces several challenges, such as the receiving and the rectifying modules of a rectenna system. Multiband antennas provide several comparative advantages against the goal of maximizing the amount of energy harvesting. In this work, we present a multiband microstrip patch antenna with three slits operating in the LoRaWAN (Long Range Wide Area Network) and the cellular (GSM-1800 and UMTS) communication frequency bands. A feasible solution of the antenna is obtained by the application of a recently introduced nature-inspired optimization technique, namely the Coyote Optimization Algorithm. The proposed antenna operates satisfactorily in the LoRaWAN (Long Range Wide Area Network) and the cellular (GSM-1800 and UMTS) communication frequency bands. Measured results of the proposed antenna exhibit an acceptable performance (multiband frequency operation, maximum gain of 3.94 dBi, broadside operation) and demonstrate features of operation, which make it a strong candidate for various RF energy harvesting applications. © 2020 IEEE.

关键词： Microstrip antennas

来源：评论

学校读者我要写书评

暂无评论

Current Issues in Flexible Manufacturing Using Multicriteria Decision Analysis and Ontology Based Interoperability in an Advanced Manufacturing Environment 10th

Current Issues in Flexible Manufacturing Using Multicriteria...

引用

10th International Conference of Production Research - Americas, ICPR-Americas 2020

作者： Canciglieri, M.B. Leite, A. F. C. S. de M. Rocha Loures, E. de F. Canciglieri, O. Monfared, R.P. Goh, Y.M. Industrial and Systems Engineering Graduate Program Pontifical Catholic University of Paraná Curitiba Paraná Brazil School of Mechanical Electrical and Manufacturing Engineering Loughborough University Loughborough Leicestershire United Kingdom

ISBN: (纸本)9783030763060

The manufacturing industry is undergoing a major transformation based on the emerging industry 4.0 technologies, such as cloud computing, big data, internet of things and cyber-physical systems. These novelty technologies aim at providing central management for the user’s flexible manufacturing requirements and information. Also, the advent of these technologies has transformed the process planning and became crucial for the building of knowledge-based process planning environments. However, current praxis cannot deal with all semantic issues within this new paradigm, as requirements must be clear, consistent, measurable, stand-alone, testable, unambiguous, unique and verifiable. In this context, multicriteria decision analysis models have gained focus of the scientific and industrial communities as a support tool for the decision-making process in the product development and advanced manufacturing as these processes excel in environments with numerous and conflicting alternatives, providing the optimal alternative. Therefore, the main objective of this research is to highlight the current issues and research tendencies regarding ontology-based interoperability systems, multicriteria decision analysis and their integration. To achieve this goal, it will be applied a literature review on the targeted technologies, discussing the current tendencies of the field and the main issues regarding their implementation and integration. Finally, the paper points themes for further research and indicates viable concepts that can compose a solution for the gaps in a systematic manner. © 2021, Springer Nature Switzerland AG.

关键词： Flexible manufacturing systems

来源：评论

学校读者我要写书评

暂无评论

Blockchain: A Review from the Perspective of Operations Researchers 22

Blockchain: A Review from the Perspective of Operations Rese...

引用

Proceedings of the Winter Simulation Conference

作者： Hong Wan Kejun Li Yining Huang Department of Industrial and Systems Engineering North Carolina State University Raleigh NC USA Operations Research Graduate Program North Carolina State University Raleigh NC USA

Blockchain is a distributed append-only digital ledger. The technology has caught much attention since the emergence of cryptocurrency, and there is an increasing number of blockchain applications in various businesses. The concept, however, is still novel to many members of the simulation and operations research community. In this tutorial, we introduce the blockchain technology and review its frontier related research. There are exciting opportunities for researchers in simulation, system analysis, and data science.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Graph Neural Networks for the Offline Nanosatellite Task Scheduling Problem

arXiv

引用

arXiv 2023年

作者： Pacheco, Bruno Machado Seman, Laio Oriel Rigo, Cezar Antônio Camponogara, Eduardo Bezerra, Eduardo Augusto dos Santos Coelho, Leandro Florianópolis Brazil Florianópolis Brazil Department of Electrical Engineering Federal University of Parana Curitiba Brazil Industrial and Systems Engineering Graduate Program Pontifical Catholic University of Parana Curitiba Brazil

This study investigates how to schedule nanosatellite tasks more efficiently using Graph Neural Networks (GNNs). In the Offline Nanosatellite Task Scheduling (ONTS) problem, the goal is to find the optimal schedule for tasks to be carried out in orbit while taking into account Quality-of-Service (QoS) considerations such as priority, minimum and maximum activation events, execution time-frames, periods, and execution windows, as well as constraints on the satellite’s power resources and the complexity of energy harvesting and management. The ONTS problem has been approached using conventional mathematical formulations and exact methods, but their applicability to challenging cases of the problem is limited. This study examines the use of GNNs in this context, which has been effectively applied to optimization problems such as the traveling salesman, scheduling, and facility placement problems. More specifically, we investigate whether GNNs can learn the complex structure of the ONTS problem with respect to feasibility and optimality of candidate solutions. Furthermore, we evaluate using GNN-based heuristic solutions to provide better solutions (w.r.t. the objective value) to the ONTS problem and reduce the optimization cost. Our experiments show that GNNs are not only able to learn feasibility and optimality for instances of the ONTS problem, but they can generalize to harder instances than those seen during training. Furthermore, the GNN-based heuristics improved the expected objective value of the best solution found under the time limit in 45%, and reduced the expected time to find a feasible solution in 35%, when compared to the SCIP (Solving Constraint Integer programs) solver in its off-the-shelf configuration. © 2023, CC BY-NC-ND.

关键词： Energy harvesting

来源：评论

学校读者我要写书评

暂无评论

Generalization Bounds for Contextual Stochastic Optimization using Kernel Regression

arXiv

引用

arXiv 2024年

作者： Wang, Yijie Hanasusanto, Grani A. Ho, Chin Pang Graduate Program in Operations Research and Industrial Engineering The University of Texas at Austin AustinTX78712-1591 United States Department of Industrial and Enterprise Systems Engineering University of Illinois Urbana-Champaign UrbanaIL61801 United States School of Data Science City University of Hong Kong Hong Kong

In this paper, we consider contextual stochastic optimization using Nadaraya-Watson kernel regression, which is one of the most common approaches in nonparametric regression. Recent studies have explored the asymptotic convergence behavior of using Nadaraya-Watson kernel regression in contextual stochastic optimization;however, the performance guarantee under finite samples remains an open question. This paper derives a finite-sample generalization bound of the Nadaraya-Watson estimator with a spherical kernel under a generic loss function. Based on the generalization bound, we further establish a suboptimality bound for the solution of the Nadaraya-Watson approximation problem relative to the optimal solution. Finally, we derive the optimal kernel bandwidth and provide a sample complexity analysis of the Nadaraya-Watson approximation problem. © 2024, CC BY.

关键词： Optimization

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：