检索结果-内蒙古大学图书馆

ieee International conference on Prognostics and Health Management (ICPHM)

作者： Ranj, Amit Guha, Arijit Routh, Bikky Pradhan, Jatin Kumar Natl Inst Technol Dept Elect Engn Rourkela 769008 Odisha India Indian Inst Technol Dept Elect Engn Kharagpur 721302 W Bengal India

ISBN: (纸本)9798350374483;9798350374476

Nowadays, Li-ion batteries (LiBs) serves as the key power source in various applications such as electronic devices, smart grid systems, electric vehicles (EVs), etc. Therefore, precise state-of-health (SOH) and end-of-life (EOL) prediction of LiBs are very much essential for their dependable and secure functioning, which can be achieved either by using a data-driven or a model-based approach. For model based approaches, the prerequisite is a precise battery model without which there can be a possibility of inaccuracies in the predictions, unlike data-driven approaches which are model free. This paper proposes a data-driven approach for the SOH prediction of LiBs based on both the capacity fading and internal resistance growth information. Accordingly, three different machine learning (ML) methods i.e., Gaussian process regression (GPR), Long short-term memory (LSTM), and Random forest (RF) have been implemented for the SOH predictions. Thereafter, a comparative analysis has been also carried out among the aforementioned methods in order to quantify the prediction accuracies. Based on the comparisons it has been observed that in the case of RF, the percentage relative error (RE) in the prediction is within 1.36% for capacity degradation and within 0.68% for internal resistance growth information which is the lowest among all the three methods corresponding to the 85th cycle with 50% of the data considered for training and the remaining data for testing purpose. Keywords: Li-ion batteries (LiBs), State-of-Health (SOH), Machine learning (ML), Gaussian process regression (GPR), Long short-term memory (LSTM), Random forest (RF), Capacity fade, Internal resistance growth.

关键词： Capacity fade Gaussian process regression (GPR) Internal resistance growth Li-ion batteries (LiBs) Long short-term memory (LSTM) Machine learning (ML) Random forest (RF) State-of-Health (SOH)

来源：评论

学校读者我要写书评

暂无评论

Camouflaged Object Detection Through Feature Selection and Enhancement 13

Camouflaged Object Detection Through Feature Selection and E...

引用

13th ieee data driven control and learning systems conference, DDCLS 2024

作者： Yuan, Yin Zhang, Kanjian Zhang, Jinxia Ministry of Education School of Automation Southeast University The Key Laboratory of Measurement and Control of Cse Nanjing210096 China Southeast University Shenzhen Research Institute Shenzhen518057 China

ISBN: (纸本)9798350361674

Camouflaged object detection is a widely explored problem, the goal of which is to completely segment objects hidden in the background. Compared with general object segmentation and salient object detection tasks, this is very challenging. Existing methods are trying their best to mine the characteristics of the camouflaged objects or obtain some auxiliary information to find the discriminative features that distinguish the targets from the background. However, these methods of explicitly supplementing information have limited exploration and analysis of camouflaged features, and modules designed for specific information are needed. The effectiveness of these modules largely affects the final results. So in this paper, an FSENet (Feature Selection and Enhancement Network) is designed based on an implicit feature selection strategy and a feature enhancement strategy. This method comprehensively utilizes the advantages of CNN and Transformer to obtain the local features and global features of the target and capture as much detail as possible. Moreover, an Implicit Feature Selection (IFS) module based on feature channels is proposed to adaptively select more important features to retain and discard unimportant feature channels to reduce interference. In addition, stacked Multi-scale Region Attention (MRA) modules are designed to enrich the multi-scale information of features and maintain the consistency of multi-scale information. Our method is evaluated on 3 challenging benchmark datasets and outperforms 20 previous state-of-the-art methods under four widely used evaluation metrics. © 2024 ieee.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

MMPF: Multimodal Purification Fusion for Automatic Depression Detection

引用

ieee TRANSACTIONS ON COMPUTATIONAL SOCIAL systems 2024年第6期11卷 7421-7434页

作者： Yang, Biao Cao, Miaomiao Zhu, Xianlin Wang, Suhong Yang, Changchun Ni, Rongrong Liu, Xiaofeng Changzhou Univ Sch Microelect & Control Engn Changzhou 213000 Jiangsu Peoples R China Changzhou Univ Sch Comp Sci & Artificial Intelligence Changzhou 213000 Jiangsu Peoples R China Soochow Univ Clin Psychol Affiliated Hosp 3 Changzhou 213000 Jiangsu Peoples R China Hohai Univ Sch Internet Things Engn Changzhou 213000 Jiangsu Peoples R China

Depression is a common mental disorder that requires objective and valid assessment tools. However, purely data-driven methods cannot satisfy the clinical diagnostic criteria for automatic depression detection (ADD), and the instability and heterogeneity of multimodal data have not been fully resolved. Therefore, we propose a novel auxiliary tool for ADD based on multimodal purification fusion (MMPF). Initially, a prior constraint gating (PCG) strategy is used to inject doctors' constraints into depression data to guide and constrain the learning process. Then, we introduce text and audio encoders to extract unpurified features from preprocessed depression data. Afterward, multimodal purification refinement is proposed to extract unintersected common and specific features from unpurified features, generating purified features. Meanwhile, we leverage a multiperspective contrastive learning (MCL) strategy to enhance unpurified and purified features. Finally, modality interaction (MI) based on the transformer is proposed to conduct multimodal fusion. A dynamic corrective learning (DCL) strategy is introduced to tackle modality imbalances and inconsistent sentiment. MMPF is evaluated on the Distress Analysis Interview Corpus Wizard of Oz and performs promisingly in unimodal and multimodal depression detection, indicating its significant role in ADD.

关键词： Automatic depression detection (ADD) contrastive learning dynamic corrective learning (DCL) multimodal fusion neural network

来源：评论

学校读者我要写书评

暂无评论

An Adaptive data-driven Iterative Feedforward Tuning Approach Based on Fast Recursive Algorithm: With Application to a Linear Motor

引用

ieee TRANSACTIONS ON INDUSTRIAL INFORMATICS 2023年第4期19卷 6160-6169页

作者： Fu, Xuewei Yang, Xiaofeng Zanchetta, Pericle Tang, Mi Liu, Yang Chen, Zhenyu Fudan Univ Sch Microelect State Key Lab ASIC & Syst Shanghai 200433 Peoples R China Univ Nottingham Dept Elect & Elect Engn Nottingham NG7 2RD England Univ Pavia Dept Elect Comp & Biomed Engn I-27100 Pavia Italy Univ Nottingham Power Elect Machine & Control Grp Nottingham NG7 2RD England Harbin Inst Technol Ctr Ultra Precis Optoelect Instrument Engn Harbin 150001 Peoples R China

The feedforward control can effectively improve the servo performance in applications with high requirements of velocity and acceleration. The iterative feedforward tuning method (IFFT) enables the possibility of both removing the need for prior knowledge of the system plant in model-based feedforward and improving the extrapolation capability for varying tasks of iterative learning control. However, most IFFT methods require to set the number of basis functions in advance, which is inconvenient to the system design. To tackle this problem, an adaptive data-driven IFFT based on a fast recursive algorithm (IFFT-FRA) is developed in this article. Explicitly, based on FRA, the proposed approach can adaptively tune the feedforward structure, which significantly increases the intelligence of the approach. Additionally, a data-based iterative tuning procedure is introduced to achieve the unbiased estimation of parameters optimization in the presence of noise. Comparative experiments on a linear motor confirm the effectiveness of the proposed approach.

关键词： Feedforward systems Tuning Iterative methods Estimation Instruments Task analysis Informatics data-based control data driven fast recursive algorithm iterative feedforward tuning (IFFT) linear motor

来源：评论

学校读者我要写书评

暂无评论

Distributed Group Coordination of Random Communication Constrained Cyber-Physical systems Using Cloud Edge Computing

IEEE TRANSACTIONS ON INDUSTRIAL CYBER-PHYSICAL SYSTEMS

引用

ieee TRANSACTIONS ON INDUSTRIAL CYBER-PHYSICAL systems 2024年 2卷 196-205页

作者： Ren, Hongru Long, Yinren Ma, Hui Li, Hongyi Guangdong Univ Technol Sch Automat Guangdong Hong Kong Joint Lab Intelligent Decis & Guangzhou 510006 Peoples R China Guangdong Univ Technol Guangdong Prov Key Lab Intelligent Decis & Coopera Guangzhou 510006 Peoples R China Guangdong Univ Technol Sch Math & Stat Guangzhou 510006 Peoples R China Southwest Univ Coll Elect & Informat Engn Chongqing 400715 Peoples R China Southwest Univ Chongqing Key Lab Gener Technol & Syst Serv Robots Chongqing 400715 Peoples R China

This paper studied the distributed group coordinated control problem of cyber-physical systems (CPSs) with multi-agent architecture. We build the distributed networked multi-group agent systems (NMGASs) with nonlinear and unknown dynamics via cloud edge computing. The common and challenging situations of random communication constraints in CPSs are considered, including network-induced delay, packet dropout, and packet disorder, which are treated as round-trip time (RTT) delay. To actively compensate for RTT delay and achieve coordination among all agents, a data-driven cloud edge predicted control strategy is designed. This strategy only needs to obtain the I/O measurement data of the systems, and can automatically carry out adaptive learning, which has more extensive application scenarios compared to model-based control methods. Theoretical analysis yields the conditions of simultaneous stability and consensus of the closed-loop systems with the proposed strategy. Finally, the practical examples are provided to illustrate the effectiveness of the proposed strategy.

关键词： Cloud computing Delays Edge computing Cyber-physical systems Predictive control Sufficient conditions Real-time systems Cloud edge computing cyber-physical systems data-driven control networked predictive control random communication constraints

来源：评论

学校读者我要写书评

暂无评论

AI-driven Educational Transformation in Secondary Schools: Leveraging data Insights for Inclusive learning Environments 15

AI-Driven Educational Transformation in Secondary Schools: L...

引用

15th ieee Global Engineering Education conference (ieee EDUCON)

作者： Duraes, Dalila Bezerra, Rita Novais, Paulo Univ Minho LASI Algoritmi Ctr Braga Portugal Polytech Cavado & Ave Inst Barcelos Portugal

ISBN: (纸本)9798350394023;9798350394030

In recent years, in the field of education, there has been a progressive trend towards teaching that is more personalised to students' characteristics and some models of prediction failure that are more accurate. In this sense, machine learning techniques have contributed to this realisation. This transformation has been significantly influenced by the integration of machine learning techniques, which have played a crucial role in harnessing data to enhance educational practices. This paper investigates the transformative potential of Artificial Intelligence (AI) within secondary education, focusing on the utilization of student assessment data and socioeconomic contextual information. The primary objective of this paper is to investigate the transformative potential of Artificial Intelligence (AI) within secondary education, with a specific focus on the utilization of student assessment data and socioeconomic contextual information. So, this paper explores the application of AI algorithms to create tailored learning pathways, adaptive support mechanisms, and targeted interventions that accommodate diverse student backgrounds. The integration of AI in secondary education is envisioned not only as a means to enhance academic outcomes but also as a tool to promote social equity and inclusivity. By leveraging data insights, educators can identify and respond to the unique needs of each student, fostering an environment where learning is optimized for individual growth. Furthermore, the paper scrutinizes the ethical considerations and challenges inherent in deploying AI systems in educational settings, emphasizing the pivotal role of equity, transparency, and data privacy in these implementations. This research aims to offer educators, policymakers, and stakeholder's insights into harnessing AI to foster adaptable, student-centric learning environments that bridge educational gaps and promote holistic academic development in secondary schools. Ethical guidelines and fra

关键词： learning Environments machine learning techniques Secondary School

来源：评论

学校读者我要写书评

暂无评论

Machine learning-Based Feature Selection for Intrusion Detection systems in IEC 61850-Based Digital Substations

Machine Learning-Based Feature Selection for Intrusion Detec...

引用

2024 International conference on Communications control and Computing Technologies for Smart Grids

作者： Eynawi, Ahmad Mumrez, Aneeqa Elhez, Ghada Hagemeyer, Veit Karlsruhe Inst Technol KIT Inst Automat & Appl Informat IAI KASTEL Secur Res Labs SRL Karlsruhe Germany

ISBN: (纸本)9798350318562;9798350318555

With the evolution in attack surfaces, smart grid complexity, and adversary models, conventional defense solutions in the smart grid domain have become less effective in providing adequate protection. In recent years, machine learning-based intrusion detection systems have been a current research topic, offering a promising and reliable defense against advanced threats. This work focuses on the development of a machine learning-based feature selection algorithm for a robust and efficient detection of attacks specifically targeting the Sampled Values and Manufacturing Message Specification protocols widely used in IEC 61850-based substations. For dataset generation, we implement several cyber-physical attacks such as injection, replay, time delay, and data modification attacks against a hardware-based substation test environment at KASTEL Security Lab Energy. The generated datasets are used to build and evaluate the developed feature selection algorithm. Our findings show that even a small number of carefully selected features can lead to more efficient and accurate anomaly detection. Considering the scarcity of datasets for smart grids, we have also released our datasets to the community for further analysis and evaluation of data-driven defense mechanisms for digital substations.

关键词： smart grids digital substations IEC 61850 Sampled Values Manufacturing Message Specification intrusion detection systems feature selection ensemble models

来源：评论

学校读者我要写书评

暂无评论

learning Robust Output control Barrier Functions From Safe Expert Demonstrations

IEEE OPEN JOURNAL OF CONTROL SYSTEMS

引用

ieee OPEN JOURNAL OF control systems 2024年 3卷 158-172页

作者： Lindemann, Lars Robey, Alexander Jiang, Lejun Das, Satyajeet Tu, Stephen Matni, Nikolai Univ Southern Calif Thomas Lord Dept Comp Sci Los Angeles CA 90089 USA Univ Penn Dept Elect & Syst Engn Philadelphia PA 19103 USA Nuro Inc Mountain View CA 94043 USA Univ Southern Calif Ming Hsieh Dept Elect & Comp Engn Los Angeles CA 90089 USA

This paper addresses learning safe output feedback control laws from partial observations of expert demonstrations. We assume that a model of the system dynamics and a state estimator are available along with corresponding error bounds, e.g., estimated from data in practice. We first propose robust output control barrier functions (ROCBFs) as a means to guarantee safety, as defined through controlled forward invariance of a safe set. We then formulate an optimization problem to learn ROCBFs from expert demonstrations that exhibit safe system behavior, e.g., data collected from a human operator or an expert controller. When the parametrization of the ROCBF is linear, then we show that, under mild assumptions, the optimization problem is convex. Along with the optimization problem, we provide verifiable conditions in terms of the density of the data, smoothness of the system model and state estimator, and the size of the error bounds that guarantee validity of the obtained ROCBF. Towards obtaining a practical control algorithm, we propose an algorithmic implementation of our theoretical framework that accounts for assumptions made in our framework in practice. We validate our algorithm in the autonomous driving simulator CARLA and demonstrate how to learn safe control laws from simulated RGB camera images.

关键词： Safety Uncertainty System dynamics Output feedback Robust control Optimization State estimation control barrier functions data-driven robust control output feedback control

来源：评论

学校读者我要写书评

暂无评论

Performance Evaluation of Support Vector Machine Algorithm in Object Classification Using Different Preprocessing Methods

Performance Evaluation of Support Vector Machine Algorithm i...

引用

ieee International conference on Automatic control and Intelligent systems (I2CACIS)

作者： Xu Zhipeng Aziz, Mohd Azri Abdul Razak, Noorfadzli Abdul Univ Teknol MARA Coll Engn Shah Alam Selangor Malaysia

ISBN: (数字)9798350372106

ISBN: (纸本)9798350372113;9798350372106

Support Vector Machines (SVMs) are excellent tools in the field of machine learning that are well suited for classification and regression problems. However, SVM is highly sensitive to data. In recent years, many researchers have continued to use various data preprocessing methods to improve model performance. Nevertheless, there is a lack of comparison of multiple datasets preprocessing methods, and there is also a lack of experiments on classification targets. This research aims to compare the performance of different preprocessing methods by using vector machines with radial basis function (RBF) kernel support. Three data sets of pedestrians, vehicles, and traffic signs are classified, and four common data preprocessing methods are used: data cleaning, feature selection, feature scaling, and dimensionality reduction. Observing the results through the use of assessment metrics accuracy, precision, recall, and F1 score. The experiment shows that the preprocessing method of feature selection is superior to the other three preprocessing methods. It achieves an accuracy of 0.985, a precision of 0.986, a recall rate of 0.985 and an F1 score of 0.986. This research has important research significance for automatic driving target classification.

关键词： SVM data preprocessing review

来源：评论

学校读者我要写书评

暂无评论

Deep Reinforcement learning for Dynamic Berth Allocation with Random Ship Arrivals 6

Deep Reinforcement Learning for Dynamic Berth Allocation wit...

引用

6th International conference on data-driven Optimization of Complex systems (DOCS)

作者： Zhou, Qianyu Wang, Peng Cao, Xiaohua Wuhan Univ Technol Sch Transportat & Logist Engn Wuhan Peoples R China

ISBN: (纸本)9798350377859;9798350377842

With the growth of global trade volume and advancements in information technology, the intelligent transformation of port systems has become a trend in the transportation industry. To address the unpredictable factors of dynamic ship arrivals, this paper utilizes a deep reinforcement learning (DRL) approach to solve the dynamic berth allocation problem (DBAP). The scheduling model aims to minimize the weighted waiting time of ships. The state space is constructed by extracting information about berths and ships in a dynamic environment. This paper proposes ship operation task buffers to map the flexible action space, and the optimization objective is decomposed by decision intervals to design the reward function. A double dueling deep Q-network (D3QN) algorithm, which integrates the advantages of DDQN and Dueling DQN, is used to solve the scheduling scheme. Finally, the network is trained with data to enable the agent to choose the optimal action based on the current state of the harbor berth system. The experimental results show that this method can effectively reduce ship waiting times in a dynamic environment, proving to be more advantageous than methods based on traditional dispatching rules.

关键词： Dynamic berth allocation Double Dueling Deep Q-Network Deep reinforcement learning Intelligent scheduling

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：