检索结果-南通市图书馆

基于深度学习的城市配电网多级动态重构决策方法

在线全文

学校读者我要写书评

暂无评论

高电压技术 2024年第4期50卷 1468-1477,I0026,I0027页

作者：姜思远高红均马望王仁浚石铖刘俊勇四川大学电气工程学院成都610065 湖北省电力公司电力科学研究院武汉430077

多级动态重构技术相比于传统全局重构技术更适合用以改善大规模城市配电网的运行经济性,但是随之而来的重构级别识别又成了新问题。由此,该文依托现有的多级动态重构数学模型,提出一种基于深度学习算法的城市配电网多级动态重构决策方法... 详细信息

多级动态重构技术相比于传统全局重构技术更适合用以改善大规模城市配电网的运行经济性,但是随之而来的重构级别识别又成了新问题。由此,该文依托现有的多级动态重构数学模型,提出一种基于深度学习算法的城市配电网多级动态重构决策方法,可跨过级别识别过程直接实现输入净负荷数据与最优重构决策方案之间的非线性映射。建立了将特征空间注意力机制、时间序列注意力机制、卷积神经网络与门控循环单元相结合的组合神经网络,针对城市配电网净负荷数据时空分布不均衡的特性,采用特征空间注意力机制与卷积神经网络对整个配电网的净负荷数据进行空间特征的学习以感知其高维特征空间的潜在联系。接着采用门控循环单元与时间序列注意力机制来充分挖掘净负荷数据在长时间尺度下的时序特征,提取其在时间分布上的不平衡特征,充分训练所提组合神经网络以拟合现有数学模型。最后,通过实际的145节点系统、IEEE 33节点系统、PG&E 69节点系统进行算例分析,验证了所提方法的有效性。

关键词：城市配电网多级动态重构时空分布不均衡组合神经网络双重注意力机制

维普期刊数据库博看期刊

计及需求响应的虚拟电厂日前市场交易策略研究

在线全文

学校读者我要写书评

暂无评论

智慧电力 2024年第7期52卷 64-71页

作者：王浩丞高红均王仁浚四川大学电气工程学院四川成都610065

虚拟电厂作为分布式资源的聚合体,可以参与电能量市场与调峰市场交易。提出一种虚拟电厂参与能量与调峰市场的交易策略。首先,构建虚拟电厂参与能量与调峰市场的交易框架。然后,构建了柔性负荷的能量-调峰响应模型,并针对柔性负荷中的... 详细信息

虚拟电厂作为分布式资源的聚合体,可以参与电能量市场与调峰市场交易。提出一种虚拟电厂参与能量与调峰市场的交易策略。首先,构建虚拟电厂参与能量与调峰市场的交易框架。然后,构建了柔性负荷的能量-调峰响应模型,并针对柔性负荷中的可转移负荷提出一种计及峰谷转移偏差的阶梯型补偿需求响应机制,目的在于引导可转移负荷参与需求响应的同时提高虚拟电厂在调峰市场中的收益。最后,构建虚拟电厂与柔性负荷之间的主从博弈模型,采用粒子群算法与Gurobi相结合的方法求解虚拟电厂利润与柔性负荷用电成本的纳什均衡。算例表明,所提需求响应机制能够充分挖掘可转移负荷的响应潜力,并增加虚拟电厂在日前市场中的利润。

关键词：虚拟电厂需求响应峰谷转移偏差日前市场

虚拟电厂两阶段准线型需求响应优化及收益共享-风险共担决策方法

在线全文

学校读者我要写书评

暂无评论

电网技术 2024年第2期48卷 799-808页

作者：成雨阳高红均王仁浚高艺文刘俊勇四川大学电气工程学院四川省成都市610065 国网四川省电力公司电力科学研究院四川省成都市610041

末端分散式资源的需求响应(demand response,DR)是支撑电力系统灵活调节的重要形式,但末端资源通常规模大且单体容量小,迫切需要虚拟电厂(virtual power plant,VPP)等市场主体进行聚合代理,从而间接参与电网公司的调节服务。聚焦于准线... 详细信息

末端分散式资源的需求响应(demand response,DR)是支撑电力系统灵活调节的重要形式,但末端资源通常规模大且单体容量小,迫切需要虚拟电厂(virtual power plant,VPP)等市场主体进行聚合代理,从而间接参与电网公司的调节服务。聚焦于准线型需求响应这一新兴的响应模式,重点研究VPP内部分散式资源的利益-风险互动协调策略,提出面向准线型DR的VPP两阶段优化及收益共享-风险共担决策方法,构建VPP与电网、内部用户互动协调新模式。首先,在准线型激励下VPP以整体收益最大化为目标,考虑柔性负荷响应积极度与新能源准确度因子,进行日前-日内两阶段DR优化;其次,在收益共享-风险共担决策方法下,VPP与柔性负荷共享准线型激励、与新能源共担不确定性风险,并以改进的Shapley法对内部柔性负荷进行利益分配。仿真结果表明,相较于传统DR,准线型DR给出全时段响应目标,更具有优越性;共享-共担决策吸引柔性负荷参与VPP调节,促使VPP与用户双赢。

关键词：虚拟电厂准线型需求响应收益共享-风险共担改进Shapley

基于一维卷积和图神经网络的配电网故障区段定位方法

同方期刊数据库博看期刊评论

在线全文

学校读者我要写书评

暂无评论

电力系统保护与控制 2024年第17期52卷 27-39页

作者：何小龙高红均黄媛高艺文王仁浚刘俊勇四川大学电气工程学院四川成都610065 国网四川省电力公司四川成都610041

快速、准确地定位故障区段对配电网的安全运行至关重要。传统故障定位方法容错率低、耗费时间长,多数深度学习算法对拓扑变动的泛化性不足。基于此,提出了一种基于一维卷积神经网络(one-dimensional convolutional neural network,1D-C... 详细信息

快速、准确地定位故障区段对配电网的安全运行至关重要。传统故障定位方法容错率低、耗费时间长,多数深度学习算法对拓扑变动的泛化性不足。基于此,提出了一种基于一维卷积神经网络(one-dimensional convolutional neural network,1D-CNN)和图神经网络(graph neural network,GNN)的配电网故障区段定位方法。该方法将配电网原始信息与GNN等深度学习算法相结合进行建模。首先利用基于注意力的时空图卷积网络从不同的时空尺度上对遥测数据进行故障特征提取,使用图注意力网络来融合多源遥信数据。然后,利用1D-CNN来调整特征输出维度以实现节点特征到故障支路的映射。最后,通过增设全连接网络来输出故障区段定位结果。依托于Matlab/Simulink平台搭建10 kV中性点不接地配电网系统进行仿真和测试。结果表明,所提方法具有优越的定位性能,能够灵活适用于各类低、中、高阻性接地故障场景,对系统拓扑变动具有强大的泛化能力以及对故障数据不完备的鲁棒性好。

关键词：配电网故障区段定位一维卷积图神经网络拓扑变动数据不完备

High-emitter identification for heavy-duty vehicles by temporal optimization LSTM and an adaptive dynamic threshold

在线全文

学校读者我要写书评

暂无评论

Frontiers of Information Technology & Electronic Engineering 2023年第11期24卷 1633-1646页

作者： Zhenyi XU renjun wang Yang CAO Yu KANG Institute of Artificial Intelligence Hefei Comprehensive National Science CenterHefei 230088China AHU-IAI AI Joint Laboratory Anhui UniversityHefei 230601China Department of Automation University of Science and Technology of ChinaHefei 230027China Institute of Advanced Technology University of Science and Technology of ChinaHefei 230088China

Heavy-duty diesel vehicles are important sources of urban nitrogen oxides(NOx)in actual applications for environmental compliance,emitting more than 80%of NOx and more than 90%of particulate matter(PM)in total vehicle *** detection and control of heavy-duty diesel emissions are critical for protecting public ***,vehicles on the road must be regularly tested,every six months or once a year,to filter out high-emission mobile sources at vehicle inspection ***,it is difficult to effectively screen high-emission vehicles in time with a long interval between annual inspections,and the fixed threshold cannot adapt to the dynamic changes of vehicle driving *** on-board diagnostic device(OBD)is installed inside the vehicle and can record the vehicle’s emission data in real *** this paper,we propose a temporal optimization long short-term memory(LSTM)and adaptive dynamic threshold approach to identify heavy-duty high-emitters by using OBD data,which can continuously track and record the emission status in real ***,a temporal optimization LSTM emission prediction model is established to solve the attention bias discrepancy problem on time steps that is caused by the large number of OBD data streams in ***,the concentration prediction error sequence is detected and distinguished from the anomalous emission contexts using flexible criteria,calculated by an adaptive dynamic threshold with changing driving ***,a similarity metric strategy for the time series is introduced to correct some pseudo anomalous *** on three real OBD time-series emission datasets demonstrate that our method can achieve high accuracy anomalous emission identification.

关键词： High-emitter identification Temporal optimization On-board diagnostic device(OBD) Dynamic threshold

MAGAN:Unsupervised Low-Light Image Enhancement Guided by Mixed-Attention

在线全文

学校读者我要写书评

暂无评论

Big Data Mining and Analytics 2022年第2期5卷 110-119页

作者： renjun wang Bin Jiang Chao Yang Qiao Li Bolin Zhang the College of Computer Science and Electronic Engineering Hunan UniversityChangsha 410006China

Most learning-based low-light image enhancement methods typically suffer from two ***,they require a large amount of paired data for training,which are difficult to acquire in most ***,in the process of enhancement,image noise is difficult to be removed and may even be *** other words,performing denoising and illumination enhancement at the same time is *** an alternative to supervised learning strategies that use a large amount of paired data,as presented in previous work,this paper presents an mixed-attention guided generative adversarial network called MAGAN for low-light image enhancement in a fully unsupervised *** introduce a mixed-attention module layer,which can model the relationship between each pixel and feature of the *** this way,our network can enhance a low-light image and remove its noise *** addition,we conduct extensive experiments on paired and no-reference datasets to show the superiority of our method in enhancing low-light images.

关键词： low-light image enhancement unsupervised learning Generative Adversarial Network(GAN) mixedattention

结合变异系数法和机器学习模型的棉花长势监测

在线全文

学校读者我要写书评

暂无评论

测绘通报 2024年第7期 111-116页

作者：杨思佳王仁军郑江华赵鹏玉韩万强毛旭芮范宏新疆大学地理与遥感科学学院新疆乌鲁木齐830046 新疆大学绿洲重点实验室新疆乌鲁木齐830046

为了更加准确地获取棉花关键物候期的长势信息,本文首先通过棉花制图指数提取棉花种植区域;然后利用变异系数法将反映棉花长势的株高、SPAD值、叶片湿重、叶片干重与叶面积5种指标构建为一个综合长势指标,即棉花长势指数(FBCGI);最后选... 详细信息

为了更加准确地获取棉花关键物候期的长势信息,本文首先通过棉花制图指数提取棉花种植区域;然后利用变异系数法将反映棉花长势的株高、SPAD值、叶片湿重、叶片干重与叶面积5种指标构建为一个综合长势指标,即棉花长势指数(FBCGI);最后选取最优特征变量,结合随机森林模型构建棉花长势反演模型。结果表明:①棉花总体分类精度达到81.65%;②与5种单一长势指标相比,构建的FBCGI与植被指数的相关性更高;③基于最优特征变量和随机森林模型构建的棉花长势监测模型,在建模集和验证集中的R 2和RMSE分别为0.74、0.07和0.51、0.10。研究结果可为棉花长势监测提供重要参考。

关键词：棉花棉花制图指数综合长势监测遥感

Microbial community coexisting with harmful alga Karenia mikimotoi and microbial control of algal bloom in laboratory

在线全文

学校读者我要写书评

暂无评论

Journal of Oceanology and Limnology 2022年第3期40卷 1027-1038页

作者： Li SUN Peike GAO Yu LI Chao wang Ning DING Junfeng CHEN Yuhao SONG Chunchen LIU Lun SONG renjun wang College of Life Sciences Qufu Normal UniversityQufu 273165China Key Laboratory of Marine Biological Resources and Ecology Liaoning Ocean and Fisheries Science Research InstituteDalian 116023China

Algicidal bacteria have been frequently isolated from algal blooming ***,knowledge regarding the microbial communities coexisting with microalgae and their potential application in preventing harmful algal blooms(HABs)is *** this study,we investigated the composition of the microbial community coexisting with harmful alga Karenia mikimotoi and its responses to algal control via nutrient stimulation or by adding algicidal strain in *** microorganisms inhabiting the *** culture consisted of 24 identifi ed phyla,including dominant Proteobacteria(relative abundance 76.24%±7.28%)and Bacteroidetes(22.67%±8.32%).Rhodobacteraceae,Phaeodactylibacter,and Maritimibacter predominated during the algal *** the added nutrient and fermentation broth of algicidal strain Pseudoalteromonas QF1 caused a massive death of *** and substantial changes in the coexisting microbial community,in which Rhodobacteraceae and Phaeodactylibacter signifi cantly decreased,while Halomonas and Alteromonas *** operational taxonomic units(OTUs)analysis indicated that 13 OTUs belonging to Rhodobacteraceae,Maritimibacter,Marivita,Nisaea,Phaeodactylibacter,Citreicella,Halomonas,Alteromonas,Marinobacter,Muricauda,and Pseudoalteromonas dominated the changes of the microbial communities observed in the *** culture with or without ***,this study indicated that microbial community inhabiting *** culture includes potential algicidal bacteria,and improves our knowledge about microbial community succession during biocontrol of *** via nutrient stimulation or by adding isolated algicidal strains.

关键词： Karenia mikimotoi microbial community nutrient stimulation algicidal bacteria Pseudoalteromonas

Decoding bone-inspired and cell-instructive cues of scaffolds for bone tissue engineering

在线全文

学校读者我要写书评

暂无评论

Engineered Regeneration 2024年第1期5卷 21-44页

作者： Zahid Hussain Shah Mehmood Xingzhu Liu Yuanshan Liu Guocheng wang renjun Pei School of Nano-Tech and Nano-Bionics University of Science and Technology of China(USTC)Hefei 230026PR China CAS Key Laboratory for Nano-Bio Interface Suzhou Institute of Nano-Tech and Nano-Bionics(SINANO)Chinese Academy of SciencesSuzhou 215123PR China Research Center for Human Tissues and Organs Degeneration Shenzhen Institute of Advanced Technology(SIAT)Chinese Academy of ScienceShenzhen 518055PR China

Bone fractures are common occurrence in clinical settings,creating a high demand for effective repair ***,limited graft availability,donor site morbidities,unpredictable clinical outcomes,immunologic reactions,infection risks,and geometrical mismatching concerns hampered tissue graft use and underscored the need for scaffolds for more effective bone reconstructions due to their tunable *** progress has been carried out in past decade in the fields of nanoceramics synthesis,bioconjugate chemistry,and composite material *** review outlines hierarchical structures and biology of bone tissue,materialistic compo-nents of scaffolds(bioceramics,polymers,bioactive drugs),featured scaffolding strategies(nanofibers,hydrogels,aerogels,bioprinting,and fiber-reinforced composite),and emphasis that hierarchical and physiochemical char-acteristics of bone should be used as an inspiration for scaffold *** review discussed how differences in materiobiological aspects of scaffolds,such as polymer/bioceramic nanocomposite,mineralized nanocomposite,matrix-rich nanocomposite,3D microenvironmental cues,pore space cues,mechanical cues,usage of physical stimulation(magnetic,electroactive,and photoactivated cues),surface cues(wettability,roughness,textured,and surface charge),and biointerface cues(cell-biomaterial interactions,cell-selective homing,and cell regula-tory strategies)modulate cellular and biological response for bone tissue *** study further outlines the challenges and benefits of integrating materiobiological cues of scaffolds for bone tissue engineering.

关键词： Bone substitute Scaffolds Mineralization Materiobiology Bone tissue engineering