检索结果-内蒙古大学图书馆

DeepOCL:A deep neural network for Object Constraint Language generation from unrestricted nature language

CAAI Transactions on Intelligence Technology 2024年第1期9卷 250-263页

作者： Yilong Yang Yibo Liu Tianshu Bao Weiru Wang Nan Niu Yongfeng Yin School of Software Beihang UniversityBeijingChina College of Computer Science and Technology Guizhou UniversityGuiyangGuizhouChina Faculty of Information Technology Beijing University of TechnologyBeijingChina Department of Electrical Engineering and Computer Sciences University of CincinnatiCincinnatiOhioUSA

Object Constraint Language(OCL)is one kind of lightweight formal specification,which is widely used for software verification and validation in NASA and Object Management Group *** OCL provides a simple expressive syntax,it is hard for the developers to write correctly due to lacking knowledge of the mathematical foundations of the first-order logic,which is approximately half accurate at the first stage of devel-opment.A deep neural network named DeepOCL is proposed,which takes the unre-stricted natural language as inputs and automatically outputs the best-scored OCL candidates without requiring a domain conceptual model that is compulsively required in existing rule-based generation *** demonstrate the validity of our proposed approach,ablation experiments were conducted on a new sentence-aligned dataset named *** experiments show that the proposed DeepOCL can achieve state of the art for OCL statement generation,scored 74.30 on BLEU,and greatly outperformed experienced developers by 35.19%.The proposed approach is the first deep learning approach to generate the OCL expression from the natural *** can be further developed as a CASE tool for the software industry.

关键词： deep learning OCL software engineering

来源：评论

学校读者我要写书评

暂无评论

A New Half-Bridge/Dual-Stacked-Switches Structured Electrolytic Capacitor-less AC/DC Bi-Directional On-Boad Charger for High-Voltage EV Battery

A New Half-Bridge/Dual-Stacked-Switches Structured Electroly...

引用

2024 IEEE Energy Conversion Congress and Exposition, ECCE 2024

作者： Derakhshan, Siamak Lam, John York University Dept. Electrical Engineering and Computer Science Toronto Canada

ISBN: (纸本)9798350376067

This paper introduces a novel half-bridge (HB)/dual-stacked-switches based electrolytic capacitor-less bidirectional AC/DC converter for high voltage (HV) electric vehicle (EV) systems. The proposed two-stage AC/DC converter incorporates a closed-loop Power Factor Correction (PFC) mechanism based on the dq model of the input LCL filter, ensuring high-quality input current with minimized Total Harmonic Distortion (THD) during both conversion and inversion operations of the converter. Moreover, the front-end HB PFC circuit operates in Continuous Conduction Mode (CCM) that results in the elimination of the high peak current in the switches. Additionally, a dedicated closed-loop control system is deployed in the dual-stacked-switches resonant converter to significantly reduce the output voltage ripple, thus eliminating the use of bulky electrolytic-type storage capacitors. Output voltage regulation is realized by using Variable Frequency Modulation (VFM). Soft-switching operation is guaranteed for all the switches and diodes using CLLC resonant circuit for both conversion and inversion modes of operation. The proposed circuit's performance is verified through a 1kW, 120Vrms/800Vdc proof-of-concept prototype. © 2024 IEEE.

关键词： Electrolytic capacitors

来源：评论

学校读者我要写书评

暂无评论

A Photovoltaic Interleaved Boost Integrated CL Resonant Converter for Magnetron Application

A Photovoltaic Interleaved Boost Integrated CL Resonant Conv...

引用

2024 IEEE Energy Conversion Congress and Exposition, ECCE 2024

作者： Molaei, Aref Kanathipan, Kajanan Lam, John York University Dept. of Electrical Engineering and Computer Science Toronto Canada

ISBN: (纸本)9798350376067

A new single-stage high-gain interleaved boost modular CL resonant converter is proposed in this paper for photovoltaic-powered magnetron application. The proposed converter is able to provide high voltage gain for the magnetron which is a device for microwave applications. Perturb and Observe (P&O) method was used to achieve Maximum Power Point Tracking (MPPT) in the integrated interleaved boost stage by controlling the duty-cycle of the two primary leg switches. The operational principles of the converter are briefly discussed in this paper. Results on a 1.7kW, 230V/3.8kV design are presented to verify the proposed converter performance. © 2024 IEEE.

关键词： Magnetrons

来源：评论

学校读者我要写书评

暂无评论

Adversarial Machine Learning Attack on Machine Learning-based Controller for Solar Inverters

Adversarial Machine Learning Attack on Machine Learning-base...

引用

2024 IEEE Energy Conversion Congress and Exposition, ECCE 2024

作者： Massa, Joaquin Kim, Kyoung-Tak Kim, Taesic Ahn, BoHyun Venayagamoorthy, Kumar Zeng, Jianwu Texas A&m University-Kingsville Dept. of Electrical Engineering and Computer Science Kingsville United States Soongsil University Dept. of Electrical Engineering Seoul Korea Republic of University of Missour-Columbia Dept. of Electrical Engineering and Computer Science Columbia United States Clemson University Dept. of Electrical and Computer Engineering Clemson United States Minnesota State University Mankato Dept. of Electrical and Computer Engineering and Technology Mankato United States

ISBN: (纸本)9798350376067

Adversarial machine learning (ML) attacks are stealthy attacks designed to mislead the ML model results. This paper explores adversarial ML attacks that generate adversarial noisy input data in an ML-based controller in a solar inverter. Three types of ML models, long short-term memory (LSTM), gated recurrent unit (GRU)), and bidirectional-LSTM (Bi-LSTM), are designed to replace proportional-integral (PI) controller-based vector control for a solar inverter and two white-box adversarial ML attacks (Basic Iterative Method (BIM) attack and Fast Sign-Gradient Method (FGSM)) are applied to the ML controllers. It is observed that the adversary ML attacks designed in stealthy way do not affect the PI controller, while significantly degrading performance of the ML-based controllers. Moreover, the BIM attack is more effective than FGSM and Bi-LSTM-based controller is relatively robust to the attacks compared to peer. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Prompt-Based Domain Incremental Learning with Modular Classification Layer 27

Prompt-Based Domain Incremental Learning with Modular Classi...

引用

27th European Conference on Artificial Intelligence, ECAI 2024

作者： Wang, Boyu Ma, Yue Qiu, Qinru Dept. of Electrical Engineering and Computer Science Syracuse University United States

ISBN: (纸本)9781643685489

Continual learning is referred to as machine learning model's ability to learn a sequence of tasks or data over time without forgetting previously learned knowledge. In particular, we focus on domain incremental learning where the model trained on one domain, e.g., real-life photos, has to be augmented to accommodate data from new domains, e.g., cartoon images. Typical incremental learning relies on rehearsal-based methods, which store trained samples in a buffer and replay them during training alongside new data. This results in significant memory overhead and raise concerns about data privacy. Recently, prompt-based methods address these challenges and outperform them by utilizing the pre-trained Vision Transformer (ViT) and replacing the replay buffer with a prompt pool. However, existing prompt-based models fail to capture domain-specific knowledge and perform poorly in domain incremental learning. In this paper, we propose "domain prompt incremental learning via dynamic neural network", which combines the advantages of architecture-based and prompt-based methods. Specifically, our framework maintains two types of prompt: a instance-level prompt that improves the model's generalization ability is shared across all input samples;and a domain prompt that encodes domain-specific knowledge is assigned for each task. Furthermore, a separated classification head is trained for each domain so that the model has a pre-trained ViT and an ensemble of classification layers, one for each domain. The experimental results shows that our approach outperforms state-of-the-art methods by 2.3% in average on two domain incremental learning (DIL) benchmarks. © 2024 The Authors.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Combined Volt-VAR and Volt-Watt Control for PCC Voltage stabilization in Smart Inverters

引用

Transactions of the Korean Institute of electrical Engineers 2023年第6期72卷 739-746页

作者： Won, Dong-Ho Lee, Kyo-Beum Dept. of Electrical and Electronic Engineering Ajou University Korea Republic of Dept. of Electrical Computer Engineering Ajou University Korea Republic of

This paper proposes the combined Volt-VAR and Volt-Watt control method of smart inverters to stabilize the point of common coupling (PCC) voltage. The distributed energy sources (DER) using renewable energy such as solar and wind power are increasing because of environmental issues worldwide. When the DER increases, the PCC voltage fluctuates, which causes a decrease in grid stability and power quality. Smart inverter provides grid support function such as Volt-VAR and Volt-Watt control method to eliminate these problems. The Volt-VAR control method adjusts capacitive/inductive reactive power based on the PCC voltage. The Volt-Watt control method stabilizes the PCC voltage by reducing active power. The combined Volt-VAR and Volt-Watt control method reduces the active power while it controls the inductive reactive power to the maximum. This paper proposes the combined Volt-VAR and Volt-Watt control method to improve the stabilization effect of PCC voltage. The effectiveness of proposed method is verified by the simulation and experiment results. Copyright © The Korean Institute of electrical Engineers.

关键词： Power quality

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Assisted Intelligent Human computer Interaction for Next Generation Internet Applications

Informatica (Slovenia)

引用

Informatica (Slovenia) 2024年第2期48卷 145-146页

作者： Hasan, Al Mehedi Hwang, Yong Seok Shin, Jungpil Dept. of Computer Science and Engineering Rajshahi University of Engineering and Technology Bangladesh Kwangwoon University Seoul Korea Republic of School of Computer Science and Engineering The University of Aizu Aizuwakamatsu Japan

来源：评论

学校读者我要写书评

暂无评论

Revitalizing the single batch environment: a ‘Quest’ to achieve fairness and efficiency

引用

International Journal of computers and Applications 2024年第8期46卷 651-665页

作者： Manna, Supriya Mudigonda, Krishna Siva Prasad Dept. of Computer Science and Engineering SRM University AP Mangalagiri Andhra Pradesh India

In the realm of computer systems, efficient utilization of the CPU (Central Processing Unit) has always been a paramount concern. Researchers and engineers have long sought ways to optimize process execution on the CPU, leading to the emergence of CPU scheduling as a field of study. In this research, we have analyzed the single offline batch processing and investigated other sophisticated paradigms such as time-sharing operating systems and wildly used algorithms, and their shortcomings. Our work is directed toward two fundamental aspects of scheduling: efficiency and fairness. We propose a novel algorithm for batch processing that operates on a preemptive model, dynamically assigning priorities based on a robust ratio, employing a dynamic time slice, and utilizing periodic sorting to achieve fairness. By engineering this responsive and fair model, the proposed algorithm strikes a delicate balance between efficiency and fairness, providing an optimized solution for batch scheduling while ensuring system responsiveness. © 2024 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Improved Object Detection Through a Learning-Based Joint-Initiative Supervised Autonomy (JISA) Framework

Improved Object Detection Through a Learning-Based Joint-Ini...

引用

2025 International Conference on Control, Automation, and Instrumentation, IC2AI 2025

作者： Tarabay, Andrea Daher, Naseem American University of Beirut Dept. of Electrical and Computer Engineering Beirut Lebanon American University of Beirut Dept. of Electrical and Computer Engineering Dept. of Mechanical Engineering Beirut Lebanon

ISBN: (纸本)9798331529604

Object detection is faced with accuracy and robustness limitations due to unstructured environments with objects of different sizes and occlusions. Despite advancements in object detection models, these models have not reached a point to possess a comparable level of perception to humans, thus limiting their ability to effectively address challenges in diverse environments. Hence, adopting a human-in-the-loop approach in autonomous systems can handle complex perception scenarios. To tackle these issues, the Joint-Initiative Supervised Autonomy (JISA) framework is used to allow a human supervisor to intervene once the agent is uncertain of its predictions based on its self-confidence (SC) metric, or when the human deems the prediction incorrect based on his/her situation awareness (SA). However, the existing JISA framework formally defines the levels of autonomy between the human and the agent, but does not provide a means for learning from human interventions. In this paper, we present an enhanced version of JISA in object detection applications by implementing a learning scheme that is inspired by the Interactive Corrections and Reinforcement for an Epistemic and Aleatoric uncertainty-aware Teaching (ICREATe) algorithm. To address these challenges, we evaluate the results of the proposed framework and demonstrate that the model effectively learns and improves its performance through human feedback. © 2025 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Enhancing the Kubernetes Platform with a Load-Aware Orchestration Strategy

引用

SN computer science 2025年第3期6卷 1-15页

作者： Marchese, Angelo Tomarchio, Orazio Dept. of Electrical Electronic and Computer Engineering University of Catania Catania Italy

Microservice architecture is increasingly favored for building large-scale applications designed for deployment in distributed and resource-constrained cloud-to-edge computing environments. As a cloud-native approach, microservices are particularly powerful due to their loosely coupled, independently deployable, and scalable characteristics. These features enable distributed deployment and flexible integration across robust cloud data centers and heterogeneous, often resource-limited, edge nodes. To fully leverage these advantages, there is a need for advanced placement algorithms that can optimize application performance by utilizing the strengths of microservices. In response to these challenges, we propose to extend Kubernetes with a load-aware orchestration strategy, which enhances its capability to deploy microservice applications within shared clusters characterized by dynamic resource usage patterns. Our approach orchestrates applications dynamically, based on real-time resource usage, allowing for continuous adjustment of their placement to match evolving conditions. Evaluation of a prototype system in a testbed environment demonstrates that this approach offers substantial benefits compared to the standard Kubernetes scheduler. © The Author(s) 2025.

关键词： Cloud continuum Container technology Kubernetes descheduler Kubernetes scheduler Microservice architecture

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：