检索结果-内蒙古大学图书馆

Energy, Electrical and Power engineering (CEEPE), International Conference on

作者： Yue Yao Ying Wang Jiangfeng Zhang Kaifeng Zhang Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Southeast University Nanjing China State Grid Zhejiang Electric Power Research Institute Zhejiang China

Accurate estimation for state-of-health (SOH) of lithium-ion batteries is critical to ensure the security and durability of battery operation. Accordingly, this paper proposes a SOH estimation method based on Attention-LSTM neural network using the partial charging features. First, features are extracted from partial constant-current charging curves. Second, Pearson coefficient analysis is employed to analysis the mapping relationship between extracted features and SOH. Third, an Attention-LSTM network is utilized to realize the SOH estimation. The attention mechanism can adaptively assign weights to the input features and time steps, thereby improving the estimation accuracy. Validation experiments are conducted on public dataset from NASA Ames Prognostics Center of Excellence. The experimental results highlight that the proposed method has higher accuracy and robustness compared with other approaches.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Double wavelet threshold filter design for real-time adaptive denoising in OA-ICOS

Double wavelet threshold filter design for real-time adaptiv...

引用

Jiangsu Annual Conference on Automation (JACA 2024)

作者： Yixin Deng Zairan Wang Yang-Yang Chen Qin Liu Xiangyu Wang School of Automation and Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Southeast University Nanjing People's Republic of China School of Cyber Science and Engineering Southeast University Nanjing People's Republic of China

ISBN: (数字)9781837242719

In the Off-Axis Integrated Cavity Output Spectroscopy (OA-ICOS) technique, single-stage filters are commonly used to redu.e noise and improve the signal-to-noise ratio, enabling more precise CO2 concentration detection. However, these filters often underperform under real-time conditions. To enhance real-time performance, this paper proposes a novel framework based on Double Wavelet Threshold Filter (DWTF) that adaptively suppresses noise while preserving key features of CO2 concentration signal. Firstly, one wavelet threshold filter (WT) is introduced to calculate the mean of the current sequence, providing an initial denoising step. Secondly, another WT is employed to extract features from a longer sequence for adaptive computation of the discount factor. Finally, an exponential moving average filter (EMA) is applied to combine both current and historical data to produce a smoother aggregated output. To validate the effectiveness, DWTF was compared with other conventional algorithms for a CO2 dataset with three distinct states: smooth, ship and mutant, obtained by OA-ICOS. The experimental results demonstrate that DWTF performs effectively in all three states.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Adaptive Neural Network Asymptotic Tracking control for Autonomous Surface Vehicles

Adaptive Neural Network Asymptotic Tracking Control for Auto...

引用

Industrial Cyber-Physical systems (ICPS)

作者： Yongchao Liu Qingzhi Wang Baozeng Fu Institute of Complexity Science School of Automation Qingdao University Qingdao China Shandong Key Laboratory of Industrial Control Technology Qingdao China Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Southeast University Nanjing China

In this paper, an adaptive neural network asymptotic tracking control method is presented for autonomous surface vehicles (ASV) with unknown uncertainties. In the control design, a smooth function is integrated into backstepping construction, which can achieve asymptotic convergence. The neural networks are borrowed to approximate the lumped nonlinear functions encompassing the unknown dynamics. It can be proved that the tracking errors of ASV can asymptotically converge to zero and all signals of the ASV closed-loop system are bounded. We offer simulation images to exhibit the validity of the devised asymptotic control method of ASV.

关键词： Adaptation models Backstepping Uncertainty Adaptive systems control design Process control Artificial neural networks

来源：评论

学校读者我要写书评

暂无评论

Close-range Hot Spot Detection based on K-neighbor Algorithm and Experiment on Hot Spot Effect 4

Close-range Hot Spot Detection based on K-neighbor Algorithm...

引用

2022 4th International Conference on Energy systems and Electrical Power, ICESEP 2022

作者： Na, ZhiXiong Wang, Dong Wang, ZiYue Wang, Jian Tang, Yu Zhang, ZiHan State Grid e-commerce Co. Ltd Beijing100053 China State Grid Tianjin Electric Power Company China Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education School of Automation Southeast University Nanjing210096 China

Abnormalities caused by the hot spot effect are very common in the application of photovoltaic modules, which seriously affect the life and power generation efficiency of photovoltaic modules. Therefore, the study of hot spot effect is of great significance to the development of photovoltaic industry in the future. By analyzing the hot spot effect and its causes, this paper lists possible factors that may affect the hot spot effect, uses the K-proximity algorithm for hot spot detection to classify the near-field infrared images obtained in photovoltaic power plants, and designs experiments to explore the influence of several factors on the hot spot effect, the influence characteristics of the shading area and the types of shading objects are deeply explored, and finally a certain summary is made. © 2022 Institute of Physics Publishing. All rights reserved.

关键词： Infrared imaging

来源：评论

学校读者我要写书评

暂无评论

A Segmented Sliding Window-Based Comprehensive Periodic Feature Extraction Method for APT Classification

A Segmented Sliding Window-Based Comprehensive Periodic Feat...

引用

International Conference on control and Automation (ICCA)

作者： Wei Song Tiejun Wu Ya Zhang School of Software Southeast University Nanjing China Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Nanjing China School of Cyber Science and Engineering Southeast University. School of Automation Southeast University Nanjing China

ISBN: (数字)9798350354409

ISBN: (纸本)9798350354416

This paper studies the classification problem of Advanced Persistent Threat (APT) attacks and proposes a comprehensive periodic feature extraction method based on sliding window division. A new data processing approach is proposed. Considering the characteristics of APT attacks, the life cycle of APT traffic data is divided into three stages. The traffic data from each stage is sampled in a piecewise manner and divided into separate training and testing sets, and then the training and testing subsets are merged. For the integrated sets, a sliding window-based partition approach is employed to obtain training and testing window samples, while extracting features from each individual sample. Then, this paper proposes a differential-based periodic feature extraction methodology to capture the dynamic information present in APT traffic and integrate it with fundamental static features, thereby forming effective comprehensive features. After combining these static and differential features, a comprehensive foundation is established for model training. Finally, the effectiveness of the algorithm is verified using decision tree model, and its superiority is demonstrated by comparing it with traditional partition methods based on data streams.

关键词： Training Heuristic algorithms Feature extraction Data models Partitioning algorithms Classification algorithms Decision trees

来源：评论

学校读者我要写书评

暂无评论

Multi-object Tracking by Detection and Query: an efficient end-to-end manner

arXiv

引用

arXiv 2024年

作者： Jia, Shukun Cao, Yichao Yang, Feng Lu, Xin Lu, Xiaobo School of Automation Southeast University Nanjing210096 China Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Southeast University Nanjing210096 China School of Instrument Science and Engineering Southeast University Nanjing210096 China

Multi-object tracking is advancing through two dominant paradigms: traditional tracking by detection and newly emerging tracking by query. In this work, we fuse them together and propose the tracking-by-detection-and-query paradigm, which is achieved by a Learnable Associator. Specifically, the basic information interaction module and the content-position alignment module are proposed for thorough information Interaction among object queries. Tracking results are directly Decoded from these queries. Hence, we name the method as LAID. Compared to tracking-by-query models, LAID achieves competitive tracking accuracy with notably higher training efficiency. With regard to tracking-by-detection methods, experimental results on DanceTrack show that LAID significantly surpasses the state-of-the-art heuristic method by 3.9% on HOTA metric and 6.1% on IDF1 metric. On SportsMOT, LAID also achieves the best score on HOTA metric. By holding low training cost, strong tracking capabilities, and an elegant end-to-end approach all at once, LAID presents a forward-looking direction for the field. Copyright © 2024, The Authors. All rights reserved.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Adapting Human Mesh Recovery with Vision-Language Feedback

arXiv

引用

arXiv 2025年

作者： Xu, Chongyang Huang, Buzhen Zhang, Chengfang Feng, Ziliang Wang, Yangang College of Computer Science Sichuan University Chengdu610065 China Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education The School of Automation Southeast University Nanjing210096 China Intelligent Policing Key Laboratory of Sichuan Province Sichuan Police College Luzhou646000 China

Human mesh recovery can be approached using either regression-based or optimization-based methods. Regression models achieve high pose accuracy but struggle with model-to-image alignment due to the lack of explicit 2D-3D correspondences. In contrast, optimization-based methods align 3D models to 2D observations but are prone to local minima and depth ambiguity. In this work, we leverage large vision-language models (VLMs) to generate interactive body part descriptions, which serve as implicit constraints to enhance 3D perception and limit the optimization space. Specifically, we formulate monocular human mesh recovery as a distribution adaptation task by integrating both 2D observations and language descriptions. To bridge the gap between text and 3D pose signals, we first train a text encoder and a pose VQ-VAE, aligning texts to body poses in a shared latent space using contrastive learning. Subsequently, we employ a diffusion-based framework to refine the initial parameters guided by gradients derived from both 2D observations and text descriptions. Finally, the model can produce poses with accurate 3D perception and image consistency. Experimental results on multiple benchmarks validate its effectiveness. The code will be made publicly available. © 2025, CC BY.

关键词： Mesh generation

来源：评论

学校读者我要写书评

暂无评论

A Pick-And-Throw Method for Enhancing Robotic Sorting Ability via Deep Reinforcement Learning 36

A Pick-And-Throw Method for Enhancing Robotic Sorting Abilit...

引用

36th Youth Academic Annual Conference of Chinese Association of Automation, YAC 2021

作者： Fang, Zihan Hou, Yanxu Li, Jun Southeast University Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Nanjing210096 China Shenzhen Research Institute Southsast University Shenzhen518057 China

ISBN: (纸本)9781665437127

To promote the work capacity and efficiency in a weakly structured logistics sorting scene, a pick-And-Throw method based on reinforcement learning is proposed. First, a D N-based learning algorithm is used to obtain a picking feasibility distribution map for guiding an optimal picking action. Second, Deep Deterministic Policy Gradient (DDPG) algorithm is utilized to train a throwing policy. The throwing policy outputs a throwing velocity of the end-effector of a robot to throw the picked object to a target area. In addition, a memory pool optimization algorithm is also proposed to enhance the convergence of the throwing policy. Experiments are conducted in both simulation and real scenarios and results demonstrate that the proposed method can improve the sorting efficiency and expand the sorting space significantly. © 2021 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A Parallel Algorithm for Solving General Colored Traveling Salesman Problems 36

A Parallel Algorithm for Solving General Colored Traveling S...

引用

36th Youth Academic Annual Conference of Chinese Association of Automation, YAC 2021

作者： Ding, Penghui Xu, Xiangping Li, Jun Ministry of Education Southeast University Key Laboratory of Measurement and Control of Complex Systems of Engineering Nanjing210096 China Southeast University Shenzhen Research Institute Shenzhen518057 China

ISBN: (纸本)9781665437127

A colored traveling salesman problem (CTSP) is a generalization of the well-known multiple traveling salesman problem. It utilizes colors to differentiate the accessibility of its cities to ward its salesmen. All the existing algorithms for solving CTSP are serial ones. This paper presents for the first time a parallel algorithm framework for solving general CTSP. Then, an improved variable neighborhood search algorithm is proposed by using this algorithm framework. Finally, extensive simulations are conducted in various test cases. The results show that the algorithm proposed is superior to the compared serial algorithms in terms of convergence and solution quality. © 2021 IEEE.

关键词： Traveling salesman problem

来源：评论

学校读者我要写书评

暂无评论

An Autoencoder with a Memory Module for Video Anomaly Detection 36

An Autoencoder with a Memory Module for Video Anomaly Detect...

引用

36th Youth Academic Annual Conference of Chinese Association of Automation, YAC 2021

作者： Tang, Wei Feng, Yunjian Li, Jun Southeast University Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Nanjing210096 China Southeast University Shenzhen Research Institute Shenzhen518057 China

ISBN: (纸本)9781665437127

With the rise of deep convolutional neural networks (CNNs), considerable attention has been paid to video anomaly detection (VAD). Autoencoders are a popular type of framework for VAD, and many existing VAD methods are based on it. However, these methods take an assumption of closed-world VAD, i.e., do not comprehensively consider the diversity of normal patterns. Besides, a competent CNN allows the autoencoder to reconstruct or predict abnormal video frames proficiently, resulting in missing anomalies. To mitigate these drawbacks, we propose an Autoencoder with a Memory Module (AMM) to realize video anomaly detection by predicting video frames. AMM consists of three modules: An encoder, a decoder, and a memory module. First, the consecutive frames are fed into the encoder to yield latent spatial features. Then, the features are utilized to retrieve corresponding memory items in the memory module to generate memory mapping features. Finally, the memory mapping features are adopted in the decoder for predicting the next frame. To match the queries against memory items accurately, we propose a memory triplet loss, which takes into account both size and angle discrepancies between the queries and memory items. At the training stage, AMM utilizes the memory triplet loss, a prediction loss, and multi-scale structure similarity measure. Moreover, the modes of retrieving and updating memory items are ameliorated by a scaled dot product model, which can alleviate vanishing gradient problems to a certain extent. Extensive experiments are conducted on three benchmark public datasets, and the results demonstrate the superior performance of AMM. © 2021 IEEE.

关键词： Forecasting

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：