检索结果-内蒙古大学图书馆

Conference on Computer Vision and Pattern recognition (CVPr)

作者： Shuailei Ma Yuefeng Wang Ying Wei Jiaqi Fan Thomas H. Li Hongli Liu Fanbing Lv Northeast University Shenyang China Information Technology R&D Innovation Center of Peking University School of Electronic and Computer Engineering Peking University Shenzhen Graduate School Shenzhen China Changsha Hisense Intelligent System Research Institute Co. Ltd

Open-world object detection (OWOd), as a more general and challenging goal, requires the model trained from data on known objects to detect both known and unknown objects and incrementally learn to identify these unknown objects. The existing works which employ standard detection framework and fixed pseudo-labelling mechanism $(PLM)$ have the following problems: (i) The inclusion of detecting unknown objects substantially reduces the model's ability to detect known ones. (ii) The $PLM$ does not adequately utilize the priori knowledge of inputs. (iii) The fixed selection manner of $PLM$ cannot guarantee that the model is trained in the right direction. We observe that humans subconsciously prefer to focus on all foreground objects and then identify each one in detail, rather than localize and identify a single object simultaneously, for alleviating the confusion. This motivates us to propose a novel solution called CAT: LoCalization and IdentificAtion Cascade detection Transformer which decouples the detection process via the shared decoder in the cascade decoding way. In the meanwhile, we propose the self-adaptive pseudo-labelling mechanism which combines the model-driven with input-driven $PLM$ and self-adaptively generates robust pseudo-labels for unknown objects, significantly improving the ability of CAT to retrieve unknown objects. Experiments on two benchmarks, i.e., MS-COCO and PASCAL VOC, show that our model outperforms the state-of-the-art methods. The code is publicly available at https://***/xiaomabufei/CAT.

关键词：

来源：评论

学校读者我要写书评

暂无评论

research and Application of Simplified Service Bus in New Generation dispatching Technical Support system and Heterogeneous system

Research and Application of Simplified Service Bus in New Ge...

引用

IEEE Conference on Energy Internet and Energy system Integration (EI2)

作者： Yi Qiang Wan Shupeng Wang Zhemin He Weihua Zhang Ping Zhang Feng System R&D Center NARI Group Corporation(State Grid Electric Power Research Institute) NARI Technology Co. Ltd Nanjing China State Grid Shanxi Electric Power Co. Ltd Taiyuan China System R&D Center State Key Laboratory of Smart Grid Protection and Control Nanjing China

With the rapid development of UHV AC / dC hybrid power grid, in order to cope with the large-scale access of new energy, the power market reform has been continuously promoted. Therefore, it is proposed to build a new generation of dispatching technical support system to meet the needs of future power system development and safe, stable and high-quality operation. This paper proposes a simplified service bus that supports a new generation of service-oriented architecture, and meets the deployment characteristics of cloud terminals that are cross platform, lightweight, easy to deploy, open and efficient. The simplified service bus can well meet the location independent characteristics and be easily deployed in and out of the system. The flexible protocol processing mode cooperating with the gateway can meet the needs of accessing the services of the system and external systems. At the same time, it supports multiple man-machine network and automatic switching of service network shutdown under N-1 mode, which greatly improves the availability and stability of the system, Which Provide a strong guarantee for the safe, stable and efficient operation of the new generation system.

关键词： Cloud computing Protocols Switches system integration Power system stability Logic gates dispatching

来源：评论

学校读者我要写书评

暂无评论

CAT: LoCalization and IdentificAtion Cascade detection Transformer for Open-World Object detection

arXiv

引用

arXiv 2023年

作者： Ma, Shuailei Wang, Yuefeng Wei, Ying Fan, Jiaqi Li, Thomas H. Liu, Hongli Lv, Fanbing Northeast University Shenyang China Information Technology R&D Innovation Center of Peking University China School of Electronic and Computer Engineering Peking University Shenzhen Graduate School Shenzhen China Changsha Hisense Intelligent System Research Institute Co Ltd China

Open-world object detection (OWOd), as a more general and challenging goal, requires the model trained from data on known objects to detect both known and unknown objects and incrementally learn to identify these unknown objects. The existing works which employ standard detection framework and fixed pseudo-labelling mechanism (PLM) have the following problems: (ii) The inclusion of detecting unknown objects substantially reduces the model’s ability to detect known ones. (ii) The PLM does not adequately utilize the priori knowledge of inputs. (iii) The fixed selection manner of PLM cannot guarantee that the model is trained in the right direction. We observe that humans subconsciously prefer to focus on all foreground objects and then identify each one in detail, rather than localize and identify a single object simultaneously, for alleviating the confusion. This motivates us to propose a novel solution called CAT: LoCalization and IdentificAtion Cascade detection Transformer which decouples the detection process via the shared decoder in the cascade decoding way. In the meanwhile, we propose the self-adaptive pseudo-labelling mechanism which combines the model-driven with input-driven PLM and self-adaptively generates robust pseudo-labels for unknown objects, significantly improving the ability of CAT to retrieve unknown objects. Experiments on two benchmarks, ***, MS-COCO and PASCAL VOC, show that our model outperforms the state-of-the-art methods. The code is publicly available at https://***/xiaomabufei/CAT. Copyright © 2023, 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

Clustering and Equivalent Modeling Method for direct-drive Wind Farms Based on Particle Swarm Neural Network

Clustering and Equivalent Modeling Method for Direct-Drive W...

引用

Energy, Power and Electrical Engineering (EPEE), International Conference on

作者： Liang Tu Libin Huang Haiping Guo Binjiang Hu Jingyue Zhang Hui Luo China Southern Power Grid State Key Laboratory of HVDC Electric Power Research Institute Guangzhou China National Energy Power Grid Technology R&D Centre China Southern Power Grid Guangzhou China Guangdong Provincial Key Laboratory of Intelligent Operation and Control for New Energy Power System China Southern Power Grid Guangzhou China China Southern Power Grid CSG Key Laboratory for Power System Simulation Electric Power Research Institute Guangzhou China National Energy Power Grid Technology R&D Centre Guangzhou China

ISBN: (数字)9798331518066

ISBN: (纸本)9798331518073

In the realm of large-scale integration of wind farms or clusters into power grids, the fault transient characteristics of these grids have experienced significant transformations. Unfortunately, the prevalent single-machine equivalent models are inadequate in accurately depicting the fault transient behavior of wind farms or clusters. Moreover, their simulation computations are exceedingly intensive, resulting in suboptimal simulation efficiency. To address these challenges, this paper introduces a clustering and equivalent modeling approach tailored for direct-drive wind farms, leveraging the capabilities of a particle swarm neural network. Initially, a transient model for direct-drive wind turbines is formulated, and fault characteristic influence factors are established through theoretical analysis. These factors encompass the distance between the wind turbine and the point of common coupling (PCC), the implementation of dC-side current limiting measures, wind speed, and reactive power at the outlet. Following this, the Euclidean distance of the influence factor set is computed, and the initial cluster centers of wind turbines are derived using an enhanced max-min distance method. Ultimately, an equivalent clustering model is formulated by utilizing the influence factors and initial cluster centers as the training set. Simulation outcomes reveal that, in comparison to the single-machine equivalent model, this method enhances simulation efficiency by a range of ${6 2 \%}$ to ${6 7 \%}$ and reduces the steady-state error in fault current simulation by ${0 . 8 \%}$ to ${1 . 9 \%}$.

关键词： Training Computational modeling Wind speed Current measurement Wind farms data models Wind turbines Steady-state Transient analysis Particle swarm optimization

来源：评论

学校读者我要写书评

暂无评论

A Transient Voltage Stability Preventive control Strategy for dC receiving-end Power system

A Transient Voltage Stability Preventive Control Strategy fo...

引用

IEEE International Conference on Power and Power and renewable Energy (ICPrE)

作者： Lian Tu Shuiping Zhang Hengdao Guo Bo Bao Yihua Zhu Jian Zuo China Southern Power Grid Power Dispatch and Control Center of Guangdong Power Grid Co. Ltd China China Southern Power Grid State Key Laboratory of HVDC Electric Power Research Institute China National Energy Power Grid Technology R&D Centre China Operation and Control for New Energy Power System Guangdong Provincial Key Laboratory of Intelligent China

With the increase of dC feed power and the decrease of the number of synchronous machines, receiving-end power system takes a serious threat to the transient voltage stability. Therefore, a transient voltage stability preventive control strategy for dC receiving-end power system is proposed in this paper. The whole method is decomposed into three sub-problems: transient voltage stability assessment, sensitivity computation and optimal control. A transient voltage stability index suitable for the sensitivity method is used to construct the transient voltage stability constraints. The accuracy and efficiency of the solution algorithm are improved by limiting the effective control interval of sensitivity and selection of the initial control set. The method has been applied in a provincial power grid in southern China. The results verifies the effectiveness of the proposed method.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Networked Method Applied to Sensitivity Analysis of a VSG controlled BESS system

A Networked Method Applied to Sensitivity Analysis of a VSG ...

引用

Power Engineering (ICPE), International Conference on

作者： Zhiyuan Sun Chao Luo Kun Zheng Yihua Zhu Jiawei Yu Guangxi Power Grid Electric Power Research Institute China Southern Power Grid Guangxi Key Laboratory of Intelligent Control and Maintenance of Power Equipment Nanning China State Key Laboratory of HVDC Electric Power Research Institute China Southern Power Grid Guang zhou China National Energy Power Grid Technology R&D Centre Guang zhou China Guangdong Provincial Key Laboratory of Intelligent Operation and Control for New Energy Power System Guang zhou China

ISBN: (数字)9798331531140

ISBN: (纸本)9798331531157

The grid-forming battery energy storage system (BESS) plays important role in constructing new energy system. This paper presents a frequency domain modal-based stability analysis of virtual synchronous generator (VSG) controlled BESS. Both ac and dc side interactive dynamics of the grid-tied inverter system are considered, by establishing a hybrid ac/dc impedance network. Then, the frequency domain modal analysis (FMA) method for networked converter systems is applied to the BESS system, covering the eigen-sensitivity from the node level to parameter level. Using this method, key parameters that influence the BESS system’s weakly/negative damping modes most can be better analyzed and identified. The analysis results are validated via the time-domain simulations.

关键词： Battery energy storage system Sensitivity analysis Frequency-domain analysis Modal analysis Stability analysis Synchronous generators Impedance Time-domain analysis Tuning Grid forming

来源：评论

学校读者我要写书评

暂无评论

Optimized Adaptive Three-Phase Active Inductance Considering Multiple Factors

Optimized Adaptive Three-Phase Active Inductance Considering...

引用

Power Electronics and Application Symposium (PEAS), IEEE International

作者： Longling Shen Guorong Zhu Feng Liu Haoran Wang Qian Wang School of Automation Wuhan University of Technology Wuhan China Yantai Power Supply Company State Grid Shandong Electric Power Company Yantai China R&D center Three Gorges Intelligent Control Technology Co. Ltd Wuhan China

Based on the topology of active inductance, an optimized adaptive three-phase active inductance control strategy considering multiple factors is proposed, and the proposed optimized adaptive active inductance is applied to the grid-side filtering of the variable frequency speed motor drive system to replace the passive inductor. Compared with the adaptive control based on grid-side current THd, this paper adds a dC bus current ripple adaptive control loop and a working condition judgment loop on this basis, which can not only further reduce the 100Hz and 200Hz harmonics generated under unbalanced working conditions, but also reduce the third harmonic on the grid-side, thereby further improving power quality. Through the dynamic characteristics simulation and comparison experiments of multiple control strategies, it can be seen that the optimized adaptive three-phase active inductor reduces the grid-side current THd and dC bus current ripple under different working conditions.4

关键词：

来源：评论

学校读者我要写书评

暂无评论

research on relay setting attack defense in power systems based on a three-layer optimization model

引用

Frontiers in Energy research 2024年 12卷

作者： ren, Liu Binyuan, Yan Hengdao, Guo Junrong, Liu Yihua, Zhu Yun, Fu Liang, Tu Zeyuan, Zhou State Key Laboratory of HVDC Electric Power Research Institute China Southern Power Grid Guangzhou China National Energy Power Grid Technology R&D Centre Guangzhou China Guangdong Provincial Key Laboratory of Intelligent Operation and Control for New Energy Power System Guangzhou China CSG Key Laboratory for Power System Simulation Electric Power Research Institute China Southern Power Grid Guangzhou China Information Centre of Guizhou Power Grid Co. Guizhou China

With the intelligent development of power systems, the number of relays continues to increase. differences in manufacturers, systems, and protocols lead to growing security risks. Tampering with relay settings could potentially cause power outages or even system instability. Consequently, relay settings have gradually become a key target for cyberattacks, particularly in smart grids where traditional defense measures struggle to effectively address complex and diverse attack scenarios. To address this issue, this paper proposes a three-layer optimization defense model based on game theory, designed to adapt to various attack scenarios. The core methodology of this model includes a three-layer structure: The first layer optimizes the protection level of each relay by allocating limited defense budgets. The second layer analyzes the potential system damage based on the attacker’s strategy choices. The third layer comprehensively calculates system losses to evaluate the effectiveness of defense plans. Through layer-by-layer optimization of budget allocation, the model minimizes the unsupplied energy loss caused by relay setting attacks. Compared to existing methods, this model not only improves defense effectiveness under resource constraints but also addresses multiple complex attack scenarios. Experimental results demonstrate that this model significantly enhances the system’s defense capabilities and effectively reduces the impact of attacks on system security operations. Copyright © 2024 ren, Binyuan, Hengdao, Junrong, Yihua, Yun, Liang and Zeyuan.

关键词： Budget control

来源：评论

学校读者我要写书评

暂无评论

Application of deep reinforcement learning in power system scheduling strategy simulation

Application of deep reinforcement learning in power system s...

引用

Energy, Power and Electrical Engineering (EPEE), International Conference on

作者： Binjiang Hu Libin Huang Yufeng Hu Jie Zhang Haiping Guo Yuanhong Lu Zeqi Hong Jingyue Zhang Hui Luo Guangdong Provincial Key Laboratory of Intelligent Operation and Control for New Energy Power System Guangzhou China State Key Laboratory of HVDC Electric Power Research Institute China Southern Power Grid Guangzhou China National Energy Power Grid Technology R&D Centre Guangzhou China CSG Key Laboratory for Power System Simulation Electric Power Research Institute China Southern Power Grid Guangzhou China

ISBN: (数字)9798331518066

ISBN: (纸本)9798331518073

With the increasing scale and complexity of power system, the current scheduling strategy methods often have the problems of limited adaptability and unsatisfactory results. Therefore, a power system scheduling method using dQN (deep Q Network) is introduced in this study. This method uses neural network model to learn and optimize scheduling strategy, and has strong adaptability and enhanced global optimization ability. Comparative evaluations with genetic algorithms and particle swarm algorithms on IEEE-118 and SG-126 datasets show that the proposed method has significant advantages in terms of cumulative reward and stable runtime.

关键词： Electrical engineering runtime Neural networks Power system stability deep reinforcement learning Stability analysis real-time systems Particle swarm optimization Optimization Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

A Novel Energy-Type SVG with Grid Forming control for Grid Voltage and Inertial Support

A Novel Energy-Type SVG with Grid Forming Control for Grid V...

引用

IEEE International Symposium on Power Electronics for distributed Generation systems (PEdG)

作者： Jiawei Zhang Chen Zhang Xianqiang Shi Lei Huang Xiaobing Li Xu Cai Key Laboratory of Control of Power Transmission and Conversion Shanghai Jiao Tong Unviersity Shanghai China Power System Analysis Department Sieyuan Electric Co. Ltd Shanghai China R&D Center Sieyuan Electric Co. Ltd Shanghai China

Large scale renewable power plants (e.g., wind farms) are typically equipped with the cascaded H-bridge based Static Var Generator (SVG) for voltage regulation of the Point of Common Coupling (PCC). To enhance its capability in gridfriendly integration and operation of renewable power plants, this paper proposed an energy-type Static-var-generator (ESVG) with grid forming (GFM) control characteristics. First, the energy type SVG's converter typology is presented, then a new type of grid forming control utilizing the distributed energy of SVG's dc capacitor is proposed, having two main characteristics: 1) it achieves autonomous frequency response and grid voltage support similar to the majority of GFM controls; 2) it has the self dc-voltage balancing capability, which does not require additional energy management control. The state-space model is also built to assess the small-signal stability of the ESVG system. Finally, an electromagnetic transient (EMT) simulation model of a typical grid-tied ESVG system is constructed in Matlab/Simulink, and simulation results show that the proposed control methods and associated characteristics are valid.

关键词：

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：