检索结果-内蒙古大学图书馆

49th IEEE International conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Santos, Augusto Rente, Diogo Seabra, Rui Moura, Jose M. E. Inst Telecomunicacoes IT Lisbon Portugal Carnegie Mellon Univ Dept Elect & Comp Engn Pittsburgh PA USA

ISBN: (纸本)9798350344868;9798350344851

In a networked Dynamical System (NDS), each node is a system whose dynamics are coupled with the dynamics of neighboring nodes. The global dynamics naturally builds on this network of couplings and it is often excited by a noise input with nontrivial structure. The underlying network is unknown in many applications and should be inferred from observed data. We assume: i) Partial observability-time series data is only available over a subset of the nodes;ii) Input noise-it is correlated across distinct nodes while temporally independent, i.e., it is spatially colored. We present a feasibility condition on the noise correlation structure wherein there exists a consistent network inference estimator to recover the underlying fundamental dependencies among the observed nodes. Further, we describe a structure identification algorithm that exhibits competitive performance across distinct regimes of network connectivity, observability, and noise correlation.

关键词： network Inference Complex systems Partial Observability Colored Noise Machine Learning

来源：评论

学校读者我要写书评

暂无评论

State Estimation in networked control systems with Time-Varying Delays: A Simple yet Powerful Observer Framework 63

State Estimation in Networked Control Systems with Time-Vary...

引用

63rd IEEE conference on Decision and control, CDC 2024

作者： Stanojevic, Katarina Steinberger, Martin Horn, Martin Institute of Automation and Control Graz University of Technology Graz Austria Institute of Automation and Control Graz University of Technology Christian Doppler Laboratory for Model Based Control of Complex Test Bed Systems Graz Austria

ISBN: (纸本)9798350316339

This paper presents a distinct framework for the design of an observer-based feedback control strategy for networked control systems (NCS) affected by unknown time-varying transmission delays larger than the sampling time Td. The pivotal element of the proposed strategy lies in the configuration of network nodes and a buffering mechanism which enable state estimation at the controller's side using observer strategies not initially designed for NCS. Consequently, the Luenberger-type observer can be designed independently from the utilized network, thus leading to a substantial reduction in the complexity typically associated with uncertain and time-varying NCS. Furthermore, the proposed framework allows the strategic incorporation of predicted state estimates in the controller design. This mitigates the effect of the network delay by reducing it by Td, which can significantly improve the system performance, as demonstrated in a simulation example. © 2024 IEEE.

关键词： networked control systems

来源：评论

学校读者我要写书评

暂无评论

Optimizing energy flexibility through electricity price-responsiveness and thermal load management in buildings with convective and radiant heating systems

引用

ENERGY AND BUILDINGS 2025年 331卷

作者： Maturo, Anthony Buonomano, Annamaria Athienitis, Andreas Concordia Univ Ctr Zero Energy Bldg Studies Montreal PQ Canada Univ Naples Federico II Dept Ind Engn Naples Italy

Building energy modeling is essential for designing energy-efficient and flexible buildings that seamlessly integrate with the electrical grid. This study introduces a data-driven, control-oriented methodology using Resistance-Capacitance thermal network models to accurately forecast building thermal loads. It differentiates the impacts of fast and slow dynamics associated with different heating types-radiant and/or convective. A Model Predictive control (MPC) framework optimizes coordination between the different building thermal dynamics, considering weather forecasts and price signals. The Varennes Library, a Net Zero Energy Institutional Building located in Quebec (Canada), serves as a case study for performance assessment. Validation of the developed model demonstrates its efficacy in enabling MPC to formulate effective control strategies. Findings reveal that high-mass radiant heating is strategically used before indoor setpoint variation or demand response events. Up to 70% of the building thermal load is delivered to the active envelope for off-peak heat storage and on-peak release. Conversely the ventilation heating is prioritized in proximity of the change in setpoint or grid tariff with percentages over 80%. Results show the adoption of weather clusters for generalizing the optimal control setting, highlighting their influence on thermal loads while maintaining robust ventilation and active envelope heating coordination. The comparison between the predictive control strategy and the existing rule-based control shows improvements in indoor temperature and energy flexibility. During the MPC routine, a constant price signal reduces grid stress, achieving Load Factor (LF) values up to 0.72 compared to 0.60 with rule-based control, while demand response, though critical peak pricing, optimally shifts up to 100% of the thermal load during peak price hours.

关键词： control-oriented model archetypes Model predictive control Energy flexibility Demand response

来源：评论

学校读者我要写书评

暂无评论

Prediction of the Static and Dynamic Characteristics of Subframe Rubber Bushings Using Deep Learning 24

Prediction of the Static and Dynamic Characteristics of Subf...

引用

24th International conference on control, Automation and systems

作者： Kim, Jieun Park, Seongkeun Soonchunhyang Univ Dept Convergence Secur Autombile Asan 31538 South Korea Soonchunhyang Univ Dept Smart Automobile Asan 31538 South Korea

ISBN: (纸本)9798331517939;9788993215380

Subframe rubber bushing play an important role in a vehicle's undercarriage. The subframe is the intermediate structure that sits between the body and the wheels, and is the main part to which the engine, transmission, suspension, and more are attached. Rubber bushing are used to connect this subframe to the body, and they perform a number of important functions. While rubber bushing are durable, they can become worn or damaged over time. To maintain rubber bushing, mathematical modeling and physical equations are often used to predict and evaluate the static and dynamic properties of rubber. However, the problem is that it takes a long time to predict the properties of rubber bushing. Therefore, this research uses deep learning models MLP, LSTM, and TCN to predict rubber bushing properties and compare their performance in order to reduce the prediction time. TCN performed the best.

关键词： Rubber bushing Static Dynamic MLP LSTM TCN Neural network

来源：评论

学校读者我要写书评

暂无评论

vStream IT: Video Streaming for Resource Constrained IoTs - An Optimal control Approach 16

vStream IT: Video Streaming for Resource Constrained IoTs - ...

引用

16th International conference on COMmunication systems and networkS, COMSNETS 2024

作者： Gupta, Rohit Mahendran, V. Badarla, Venkataramana Indian Institute of Technology Tirupati Department of Cse India

ISBN: (纸本)9798350383119

In the smart industrial era, with the rise of AR/VR technologies, dense deployments of IoTs such as 'head mounted displays' and 'smart glasses' are expected to become pervasive. For such dense IoT client deployments, edge-server based video streaming framework can improve the network performance such as low latency, and high and stable throughput. Such network enhancements can be utilized to improve the performance of end-to-end video (application) streaming by appropriately designing Adaptive Bit Rate (ABR) algorithm that also takes into account the resource constrained nature of the hosting IoTs. In this paper, we develop one such IoT friendly deterministic optimal control based ABR streaming algorithm. The exponential complexity of underlying state space is tackled by exploiting the underlying problem structure to yield an optimal and computationally-simple ABR algorithm namely, Video Streaming for IoTs (vStreamIT), which also use minimal storage. The efficacy of the proposed vStreamIT algorithm is evaluated against the state of the art and is shown to exhibit improved streaming performance in terms of bit-rate, while consuming less buffer space (a critical storage constraint for IoTs). © 2024 IEEE.

关键词： DASH deterministic optimal control edge networks IoT MDP performance modeling and analysis

来源：评论

学校读者我要写书评

暂无评论

Beyond Traditional Methods: Deep Learning with Data Augmentation for Robust Access control 33

Beyond Traditional Methods: Deep Learning with Data Augmenta...

引用

33rd IEEE International conference on Computer Communications and networks (ICCCN)

作者： Al Lail, Mustafa Pinto, Daniela Almanza, Luis Alvarez Salazar, Francisco Rizzi, Carolina Texas A&M Int Univ Sch Engn Laredo TX 78041 USA

ISBN: (纸本)9798350348439;9798350384611

Access control systems in large organizations often struggle with managing complex policies and workloads. However, there is potential for deep learning models to address these challenges. This study delves into the suitability of various deep learning architectures for making real-time access control decisions. Six prominent Convolutional Neural network (CNN) models (ResNet, DenseNet, Xception, Inception, AlexNet, VGG-16) are evaluated, and the impact of data augmentation using SMOTE on their performance is analyzed. The findings demonstrate that most deep learning models consistently deliver results in access control. ResNet outperforms other models, showing high accuracy across original and SMOTE-augmented datasets. Moreover, SMOTE generally enhances performance for most models, highlighting its potential for addressing data imbalance. These results indicate that deep learning shows promise for improving access control tasks.

关键词： Access control Deep Learning CNN Security

来源：评论

学校读者我要写书评

暂无评论

control-Oriented Modeling for Industrial Propylene Polymerization Process based on Physics-Informed Neural network 14

Control-Oriented Modeling for Industrial Propylene Polymeriz...

引用

14th Asian control conference (ASCC)

作者： Li, Shenjie Zhang, Xixiang Tian, Zhou Lu, Jingyi Du, Wenli Qian, Feng East China Univ Sci & Technol Minist Educ Key Lab Smart Mfg Energy Chem Proc Shanghai 200237 Peoples R China Qingyuan Innovat Lab Quanzhou 362801 Peoples R China

ISBN: (纸本)9798331540845;9789887581598

Loop reactors are extensively employed in the industrial production of polypropylene. Nonetheless, the high nonlinearity and stiffness of this process pose significant challenges. Traditional methods are time-consuming in simulating this process. This work is aimed at using Physics-Informed Neural network (PINN) to improve the computation efficiency. PINN can use neural network to directly give the solution without solving the ODEs through explicit numerical methods. Particularly, we extend the application of PINNs to include initial states and control inputs, making them suitable for control tasks. In order to enable the model to track the stiffness variables and to improve the generalization performance, we combine Latin Hypercube Sampling (LHS) Gaussian-Lagrange method for sampling time, states and control collocation points. Furthermore, we utilize the Extended Physics-Informed Neural networks framework, which ensures that solutions inherently satisfy initial conditions and constraints. The control-Oriented Physics-Informed Neural network (COPINN) proposed here achieves calculation accuracy comparable to the Backward Differentiation Formula (BDF) method while significantly improving computational efficiency.

关键词： Physics-Informed Neural network Process Modeling Stiff systems

来源：评论

学校读者我要写书评

暂无评论

Enhancing Accuracy Through Data Augmentation using Variational Autoencoders in Machine Learning Techniques

Enhancing Accuracy Through Data Augmentation using Variation...

引用

control Instrumentation System conference (CISCON) - Guiding Tomorrow - Emerging Trends in control, Instrumentation, and systems Engineering

作者： Dubey, Vaibhav Kaur, Bhavneet Goel, Paurav Chandigarh Univ Dept CSE Ajitgarh Punjab India

ISBN: (纸本)9798350375480;9798350375497

This paper investigates the impact of data augmentation on the performance of a simple Deep Neural network (DNN) architecture, comprising a sequence of fully connected layers interspersed with non-linear activation functions and regularization techniques. The model was trained over 2000 epochs, with and without data augmentation. Rectified Linear Unit (ReLU) activation functions are applied after each layer, except the last, to introduce non-linearity, while dropout regularization is used to enhance model robustness and prevent overfitting. Experimental results reveal that data augmentation significantly enhances model performance. The registered training loss decreased and training accuracy improved significantly with data augmentation technique. The same results are also observed in validation phase. These improvements suggest that data augmentation aids the model in learning more effectively by introducing variability in the training data, enhancing its generalization capability and robustness to variations in unseen datasets.

关键词： Data Augmentation Variational Auto Encoders (VAE) Convolutional Neural network (CNN) Machine Learning (ML) Feature Engineering Data Balancing

来源：评论

学校读者我要写书评

暂无评论

MPPT control for Standalone Wind Energy Conversion System: Integrating Hybrid P&O and ANN with Voltage-Lift Boost Converter 7

MPPT Control for Standalone Wind Energy Conversion System: I...

引用

7th International conference on the Development in the Renewable Energy Technology (ICDRET)

作者： Saifullah, Md. Khaled Jahan, Effat Uddin, Mohammad Nasir Amer Int Univ Bangladesh Dept Elect & Elect Engn Dhaka Bangladesh

ISBN: (纸本)9798350373394;9798350373400

This study proposes an approach to enhance the performance of wind energy conversion systems (WECS) through the integration of a hybrid Perturb and Observe (P&O) and Artificial Neural network (ANN)-based Maximum Power Point Tracking (MPPT) control technique. The application of this technique is explored in the context of WECS utilizing voltage lift-based DC/DC converters. The utilization of an artificial neural network control method aims to optimize power extraction from varying wind velocities effectively. The system incorporates a voltage lift-boost converter to achieve stable voltage and a higher power rating. Additionally, a standard boost converter is employed for comparative analysis, serving as a benchmark to validate the proposed converter's performance. The study systematically assesses and compares the output power, convergence time, and stability of three configurations: the traditional P&O technique, the hybrid MPPT method with a conventional boost converter, and the hybrid MPPT method with a voltage lift-boost converter. The study conducts a thorough analysis of the MPPT technique and DC/DC converter performance using MATLAB/Simulink.

关键词： Wind Energy Conversion System (WECS) PMSG MPPT hybrid artificial neural network (ANN) converter voltage power Voltage-Lift (VL) boost

来源：评论

学校读者我要写书评

暂无评论

Robust Adaptive Neural network-Based Funnel Tracking control of a Class of Perturbed Euler-Lagrange systems 20th

Robust Adaptive Neural Network-Based Funnel Tracking Control...

引用

20th International conference on Intelligent Computing (ICIC)

作者： Tong, Xingcheng Zhao, Zhiye Jin, Xiaozheng Qilu Univ Technol Shandong Acad Sci Minist EducNatl Supercomp Ctr Jinan Shandong Comp Sci CtrKey Lab Comp Power Network Jinan Peoples R China Shandong Fundamental Res Ctr Comp Sci Shandong Prov Key Lab Comp Networks Jinan Peoples R China

ISBN: (纸本)9789819756742;9789819756759

In this article, a robust adaptive neural network (NN)-based funnel tracking control method is investigated to address anti-perturbation issues in disturbed Euler-Lagrange (EL) systems. The NN-based approaches are used to approximate the unknown nonlinear functions, whereas adaptive techniques are employed to estimate disturbance bounds. Meanwhile, an adaptive compensation control scheme is introduced, leveraging neural networks to mitigate above effects of nonlinear functions and disturbances. Additionally, a control strategy resembling a funnel is devised to confine trajectory tracking errors within a predefined region and finite time despite persistent perturbations. The bounded stability of tracking errors in EL systems with prescribed performance is established using Lyapunov stability theory. Simulations are performed to confirm the efficacy of the control technology.

关键词： Neural network Funnel control Perturbation rejection Euler-Lagrange systems

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：