检索结果-南通市图书馆

Super-resolving microscopy images of Li-ion electrodes for fine-feature quantification using generative adversarial networks

在线全文

学校读者我要写书评

暂无评论

npj Computational Materials 2022年第1期8卷 650-660页

作者： Orkun Furat Donal P.Finegan Zhenzhen Yang Tom Kirstein Kandler Smith Volker Schmidt Institute of Stochastics Ulm UniversityHelmholtzstraße 1889069UlmGermany National Renewable Energy Laboratory 15013 Denver W ParkwayGoldenCO80401USA Chemical Sciences and Engineering Division Argonne National Laboratory9700 S.Cass AvenueLemontIL60439USA

For a deeper understanding of the functional behavior of energy materials,it is necessary to investigate their microstructure,e.g.,via imaging techniques like scanning electron microscopy (SEM).However,active materials are often heterogeneous,necessitating quantification of features over large volumes to achieve representativity which often requires reduced resolution for large fields of *** within Li-ion electrode particles are an example of fine features,representative quantification of which requires large volumes of tens of *** overcome the trade-off between the imaged volume of the material and the resolution achieved,we deploy generative adversarial networks (GAN),namely SRGANs,to super-resolve SEM images of cracked cathode materials.A quantitative analysis indicates that SRGANs outperform various other networks for crack detection within aged cathode *** makes GANs viable for performing super-resolution on microscopy images for mitigating the trade-off between resolution and field of view,thus enabling representative quantification of fine features.

关键词： microstructure crack networks

Almost Automorphic Solutions for Quaternion-Valued Hopfield Neural networks with Mixed Time-Varying Delays and Leakage Delays

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Systems Science & Complexity 2020年第1期33卷 100-121页

作者： LI Yongkun MENG Xiaofang Department of Mathematics Yunnan UniversityKunming 650091China

This paper considers a class of quaternion-valued Hopfield neural networks with mixed time-varying delays and leakage *** utilizing the exponential dichotomy of linear differential equations,Banach’s fixed point theorem and differential inequality techniques,the authors obtain some sufficient conditions to ensure the existence and global exponential stability of almost automorphic solutions for this class of quaternion-valued neural *** results are completely ***,the authors give an example to illustrate the feasibility of the results.

关键词： Almost automorphic solutions exponential stability Hopfield neural networks leakage delays quaternion-valued functions

Wireless Acoustic Sensor networks and Edge Computing for Rapid Acoustic Monitoring

在线全文

学校读者我要写书评

暂无评论

IEEE/CAA Journal of Automatica Sinica 2019年第1期6卷 64-74页

作者： Zhengguo Sheng Saskia Pfersich Alice Eldridge Jianshan Zhou Daxin Tian Victor C.M.Leung the Department of Engineering and Design University of Sussex Infineon Technologies AG the Department of Music University of Sussex Beijing Advanced Innovation Center for Big Data and Brain Computing (BDBC) Beijing Key Laboratory for Cooperative Vehicle Infrastructure Systems & Safety ControlSchool of Transportation Science and EngineeringBeihang University the Department of Electrical and Computer Engineering The University of British Columbia

Passive acoustic monitoring is emerging as a promising solution to the urgent, global need for new biodiversity assessment methods. The ecological relevance of the soundscape is increasingly recognised, and the affordability of robust hardware for remote audio recording is stimulating international interest in the potential for acoustic methods for biodiversity *** scale of the data involved requires automated methods,however, the development of acoustic sensor networks capable of sampling the soundscape across time and space and relaying the data to an accessible storage location remains a significant technical challenge, with power management at its core. Recording and transmitting large quantities of audio data is power intensive,hampering long-term deployment in remote, off-grid locations of key ecological interest. Rather than transmitting heavy audio data, in this paper, we propose a low-cost and energy efficient wireless acoustic sensor network integrated with edge computing structure for remote acoustic monitoring and in situ *** and computation of acoustic indices are carried out directly on edge devices built from low noise primo condenser microphones and Teensy microcontrollers, using internal FFT hardware support. Resultant indices are transmitted over a ZigBee-based wireless mesh network to a destination *** tests of audio quality, indices computation and power consumption demonstrate acoustic equivalence and significant power savings over current solutions.

关键词： Acoustic sensor networks edge computing energy efficiency

Virtual sensing method for monitoring vibration of continuously variable configuration structures using long short-term memory networks

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Aeronautics 2020年第1期33卷 244-254页

作者： Zhenjiang YUE Li LIU Teng LONG Yuanchen MA School of Aerospace Engineering Beijing Institute of TechnologyBeijing 100081China Key Laboratory of Dynamics and Control of Flight Vehicle Ministry of EducationBeijing 100081China

Vibration monitoring by virtual sensing methods has been well developed for linear timeinvariant structures with limited ***,few methods are proposed for Time-Varying(TV)structures which are inevitable in aerospace *** core of vibration monitoring for TV structures is to describe the TV structural dynamic characteristics with accuracy and *** paper propose a new method using the Long Short-Term Memory(LSTM)networks for Continuously Variable Configuration Structures(CVCSs),which is an important subclass of TV *** configuration parameters are used to represent the time-varying dynamic characteristics by the‘‘freezing"*** relationship between TV dynamic characteristics and vibration responses is established by LSTM,and can be generalized to estimate the responses with unknown TV processes benefiting from the time translation invariance of LSTM.A numerical example and a liquid-filled pipe experiment are used to test the performance of the proposed *** results demonstrate that the proposed method can accurately estimate the unmeasured responses for CVCSs to reveal the actual characteristics in time-domain and ***,the average one-step estimation time of responses is less than the sampling ***,the proposed method is promising to on-line estimate the important responses of TV structures.

关键词： Data-based method Recurrent neural networks Time-varying structure Vibration monitoring Virtual sensing

Approximate Data Aggregation in Sensor Equipped IoT networks

在线全文

学校读者我要写书评

暂无评论

Tsinghua Science and Technology 2020年第1期25卷 44-55页

作者： Ji Li Madhuri Siddula Xiuzhen Cheng Wei Cheng Zhi Tian Yingshu Li Kennesaw State University MariettaGA 30060USA Georgia State University AtlantaGA 30303USA George Washington University WashingtonDC 20052USA Department of Computer Science Virginia Commonwealth UniversityRichmondVA 23284USA Department of Electrical&Computer Engineering George Mason UniversityFairfaxVA 22030USA

As Internet-of-Things(IoT) networks provide efficient ways to transfer data, they are used widely in data sensing applications. These applications can further include wireless sensor networks. One of the critical problems in sensor-equipped IoT networks is to design energy efficient data aggregation algorithms that address the issues of maximum value and distinct set query. In this paper, we propose an algorithm based on uniform sampling and Bernoulli sampling to address these issues. We have provided logical proofs to show that the proposed algorithms return accurate results with a given probability. Simulation results show that these algorithms have high performance compared with a simple distributed algorithm in terms of energy consumption.

关键词： data aggregation sampling Internet-of-Things(IoT) networks

Learning Bayesian networks by constrained Bayesian estimation

在线全文

学校读者我要写书评

暂无评论

Journal of Systems Engineering and Electronics 2019年第3期30卷 511-524页

作者： GAO Xiaoguang YANG Yu GUO Zhigao School of Electronics and Information Northwestern Polytechnical University

Bayesian networks (BNs) have become increasingly popular in recent years due to their wide-ranging applications in modeling uncertain knowledge. An essential problem about discrete BNs is learning conditional probability table (CPT) parameters. If training data are sparse, purely data-driven methods often fail to learn accurate parameters. Then, expert judgments can be introduced to overcome this challenge. Parameter constraints deduced from expert judgments can cause parameter estimates to be consistent with domain knowledge. In addition, Dirichlet priors contain information that helps improve learning accuracy. This paper proposes a constrained Bayesian estimation approach to learn CPTs by incorporating constraints and Dirichlet priors. First, a posterior distribution of BN parameters is developed over a restricted parameter space based on training data and Dirichlet priors. Then, the expectation of the posterior distribution is taken as a parameter estimation. As it is difficult to directly compute the expectation for a continuous distribution with an irregular feasible domain, we apply the Monte Carlo method to approximate it. In the experiments on learning standard BNs, the proposed method outperforms competing methods. It suggests that the proposed method can facilitate solving real-world problems. Additionally, a case study of Wine data demonstrates that the proposed method achieves the highest classification accuracy.

关键词： Bayesian networks (BNs) parameter learning constraints sparse data

Task-Aware Flow Scheduling with Heterogeneous Utility Characteristics for Data Center networks

在线全文

学校读者我要写书评

暂无评论

Tsinghua Science and Technology 2019年第4期24卷 400-411页

作者： Fang Dong Xiaolin Guo Pengcheng Zhou Dian Shen the School of Computer Science and Engineering Southeast University

With the continuous enrichment of cloud services, an increasing number of applications are being deployed in data centers. These emerging applications are often communication-intensive and data-parallel, and their performance is closely related to the underlying network. With their distributed nature, the applications consist of tasks that involve a collection of parallel flows. Traditional techniques to optimize flow-level metrics are agnostic to task-level requirements, leading to poor application-level performance. In this paper, we address the heterogeneous task-level requirements of applications and propose task-aware flow scheduling. First, we model tasks' sensitivity to their completion time by utilities. Second, on the basis of Nash bargaining theory, we establish a flow scheduling model with heterogeneous utility characteristics, and analyze it using Lagrange multiplier method and KKT condition. Third, we propose two utility-aware bandwidth allocation algorithms with different practical constraints. Finally, we present Tasch, a system that enables tasks to maintain high utilities and guarantees the fairness of utilities. To demonstrate the feasibility of our system, we conduct comprehensive evaluations with realworld traffic trace. Communication stages complete up to 1.4 faster on average, task utilities increase up to 2.26,and the fairness of tasks improves up to 8.66 using Tasch in comparison to per-flow mechanisms.

关键词： data center networks coflow flow scheduling data-intensive applications

Traffic dynamics considering packet loss in finite buffer networks

在线全文

学校读者我要写书评

暂无评论

Chinese Physics B 2019年第4期28卷 433-438页

作者： Jie Chen Jin-Yong Chen Ming Li Mao-Bin Hu School of Engineering Science University of Science and Technology of China Civil Architectural and Environmental Engineering University of Texas at Austin

In real complex systems, the limited storage capacity of physical devices often results in the loss of data. We study the effect of buffer size on packet loss threshold in scale-free networks. A new order parameter is proposed to characterize the packet loss threshold. Our results show that the packet loss threshold can be optimized with a relative small buffer size. Meanwhile, a large buffer size will increase the travel time. Furthermore, we propose a Buffered-Shortest-Path-First(BSPF) queuing strategy. Compared to the traditional First-In-First-Out(FIFO) strategy, BSPF can not only increase the packet loss threshold but can also significantly decrease the travel length and travel time in both identical and heterogeneous node capacity cases. Our study will help to improve the traffic performance in finite buffer networks.

关键词： finite buffer networks loss threshold queuing strategy

A Linear Frequency Principle Model to Understand the Absence of Overfitting in Neural networks

在线全文

学校读者我要写书评

暂无评论

Chinese Physics Letters 2021年第3期38卷 121-126页

作者： Yaoyu Zhang Tao Luo Zheng Ma Zhi-Qin John Xu School of Mathematical Sciences Institute of Natural SciencesMOE-LSCand Qing Yuan Research InstituteShanghai Jiao Tong UniversityShanghai 200240China Shanghai Center for Brain Science and Brain-Inspired Technology Shanghai 200031China

Why heavily parameterized neural networks(NNs) do not overfit the data is an important long standing open question. We propose a phenomenological model of the NN training to explain this non-overfitting puzzle. Our linear frequency principle(LFP) model accounts for a key dynamical feature of NNs: they learn low frequencies first, irrespective of microscopic details. Theory based on our LFP model shows that low frequency dominance of target functions is the key condition for the non-overfitting of NNs and is verified by experiments. Furthermore,through an ideal two-layer NN, we unravel how detailed microscopic NN training dynamics statistically gives rise to an LFP model with quantitative prediction power.

关键词： networks neural details